A manual for species counterpoint

Transcription

A Manual for Species Counterpoint
Musica Practica 1
University of Wisconsin-Madison
September 2001
.O 2001 by Brian Hyer. No unauthorized use permitted.
A Manual for Species Counterpoint
Table of Contents
..................................................................................1
1.1 Introduction ................................................................................................................ 1
1.2 Performance ............................................................................................................... 7
Vocal Production ........................................................................................................
8
Gesture ........................................................................................................................8
Solfege ...................................................................................................................10
1.3 Basic Concepts ..........................................................................................................10
Cantus firmus ...........................................................................................................10
Mode .......................................................................................................................... 11
Musical Goals .......................................................................................................... 12
Consonance and Dissonance ........................
.
......................................................12
Melodic Range (Ambitus) and Distance between Voices .....................................13
Simple and Compound Intervals ........................................................................... 13
Four Varieties of Relative Motion ..........................................................................13
Clausula formalis ..................................................................................................... 15
1.4 Rules ..........................................................................................................................17
Rules Governing Relations between Voices ........................................................... 18
Rules Governing Melodic Behavior .................................................................... 20
Chapter 1 .
First Species Counterpoint
1.5 How to Compose First Species Counterpoint above a Cantus firmus
........................................................................................... 30
Common Compositional Problems ......................................................................... 33
1.6 Invertible Counterpoint
1.7
............... 23
.................................... . ....................................36
Introduction .............................................................................................................
36
Basic Concepts ............................
.
.........................................................................40
Meter ......................................................................................................................... 40
Weak Beats as Means of Transition ....................................................................
41
On the Absence of Dissonance in First Species ...................................................44
Possible Uses of the Second Half Note ..................................................................45
Conjunct Motion ....................................................................................................... 45
Disjunct Motion ........................................................................................................ 47
Rules .......................................................................................................................... 50
Chapter 2 -- Second Species Counterpoint
2.1
2.2
2.3
2.4
Rules Governing Melodic Behavior
Musical Goals
........................................................................49
........................................................................................................... 53
................................................................... 54
Apex Repetition ........................................................................................................ 55
A Case of Mistaken Musical Identities
2.5 A Second Species Counterpoint from Gradus ad Parnassum
Chapter 4 -Fourth Species Counterpoint
............................ 56
..................... ....................................................62
.......................................................................................................... 62
Dissonant Suspensions ........................................................................................66
4.1 Introduction
4.2
Writing Suspensions above a Cantus firmus
........................................................ 67
Writing Suspensions below a Cantus firmus
....................................................... 68
......................................................................................... 69
Unisons ....................................................................................................................... 71
4.3 Consonant Syncopations
...........................................................................................72
Musical Goals ...........................................................................................................
74
Rules ................................................................................................................... 75
4.4 Breaking the Species
4.5
........................................................................76
I
Chapter 1
1.1 Introduction
Ex. 1.1 presents a score for a complete musical composition. It is a remarkable piece,
for a number of reasons, one of them being the score, a glance at which suggests that the
piece is simpler than the music most of us are used to listening to, though - as will soon
become clear - the piece is far more complicated than it would appear.
Ex. 1.1
Even without listening to the piece, it is clear that it is extremely brief: there is no indication of tempo in the score, but even a t a slow pace a performance of the entire piece would
last less than ten seconds. What's more, the score includes almost no information about
instrumentation, though the mere fact that there are two staves suggests that there are
two musical parts or - in this case - voices: within the musical tradition fkom which this
piece stems it is understood that the composition is for two singers (or two groups of singers, each group singing the same part), one voice per staff.
We have enough information a t this point to approximate a performance of the music,
"approximate" because as we continue to perform, listen to, and discuss this piece, we will
extend and refine our ideas about what an appropriate or effective performance of the
music might mean. For now, sing both parts, first alone, then together with someone else;
if possible, switch parts in order to get a better, more complete sense of how the piece
goes. If no one is available to sing the other part, sing one part while playing the other on
the piano, or perform both parts on the piano. It is always preferable,, however, to sing
this music.
When singing this music, use solfege, vocalizing each pitch with the syllable shown in
Ex. 1.2:
Ex. 1.2
re
fa
mi
re
sol
fb
la
sbl
fa
mi
re
2
Chapter 1
Most college-age musicians are familiar with solfege, though few have much experience
using it. While the claims made for (and about) solfege are sometimes exaggerated, it does
have a number of practical uses. Solfege makes it easier for us to talk about what we
hear and to communicate about it with other musicians. It also makes for a more musical
performance of the piece: because each syllable begins with a consonant (re, mi, fa, etc.),
using solfege allows for a cleaner, more precise attack of each note, which in turn makes
it easier for us synchronize the two parts in performance. Solfege is also more listenable
than other alternatives. Singing each note on la, for instance, can be vocally tiring: when
singing on la, we often begin to sing toward the base of the tongue, producing an ugly,
glottal sound, as if we're swallowing the pitches. Solfege, in contrast, requires a change of
vowel - most of which require a brighter, "more forward* vocal production - with each
pitch.
Ex. 1.2 gives the appropriate solfege syllable below each note in both voices. In the
upper part, the next-to-last note - a C# - is labeled si. Without for the moment going
into detail, si is being used in this one instance as a functional designation to label the
leading tone: that is, C# is to D as B is -within the diatonic, white-note collection - to C;
just as B forms a si in relation to do, C# can be heard as a si (a leading tone) in relation to
re. From now on, use solfege when singing the examples; the syllables, however, will no
longer be provided.
A few more instructions. Even though both voices consist of whole notes, choose a
flowing, moderate tempo, not too slow, about two whole notes per second, and sing at an
even pace from beginning to end; if possible, sing the entire part in a single breath. Each
part, moreover, should be sung legato, so that there is a smooth, seamless connection
between each note and the next, and so that each note is articulated without a discernible
accent. Choose a comfortable register to sing in: in this case, basses and tenors will want
to sing the music on the upper staff two octaves below the notated pitches, while sopranos and altos will feel more comfortable singing the same music an octave lower than
written. Performance will be addressed in greater detail in Section 1.2, but for now, experiment with the performance, following your own musical instincts about how best to
realize the music.
Ex. 1.1instantiates a kind of music known as species counterpoint. In its earliest historical forms, species counterpoint arose (in the late renaissance) as a pedagogical practice
-as a means of instruction for musicians, whether performers or composers - a purpose
for which it is still being used, over four centuries later. Its pedagogical aims account in
large measure for its intentional simplification of musical texture, at least in comparison
with other contemporaneous compositional practices. Ex. 1.1consists, as noted above, of
nothing but whole notes (or semibreves) in both voices - at least, that is, until the breve
(double whole note) of the final measure - and there is a one-to-one, semibreve-to-semibreve correlation between the two voices; first species is often described, for this reason,
as one-to-one, note-against-note counterpoint.
Another way in which this piece is uncomplicated is that it draws on a limited, finite
collection of pitches. In this case, that collection consists of diatonic (or natural) pitches:
except for the accidental in the next to last (or penultimate) measure, there is no chromaticism in this piece. Moreover, stepwise melodic motion is the norm in both voices: leaps
are for the most part few and far between. And the melodic range (or ambitus) of each
voice is rather narrow: the upper voice remains within the major 6th between a low C and
a high A, while the lower voice has even narrower melodic compass, moving entirely within the perfect 5th between a low D and a high A.
3
Both voices, moreover, are about equal in melodic interest: neither voice predominates, takes precedence over, or distracts attention from the other - we hear both voices,
in other words, as individual melodies. We are of course all familiar with melodies from
other sorts of music, but one of the remarkable things about species counterpoint is that
it allows us to discuss, in considerable detail, how we hear them, but also why some melodies seem better or more effective than others.
In thinking about melodies, musicians often draw on two intersecting families of metaphors, mental images, that is, of lines and streams. When we perform and listen to melodies, we hear them less as successions of individual notes, occurring one after the other,
than as notes that connect together to form a melodic line. In this vein we imagine melodies as contours plotted on a mental graph where the vertical axis designates pitch (running from low to high) and the horizontal axis designates the lapse of musical time (measured out, in the case of Ex. 1.1,in whole notes). All melodies, in other words, emerge over
time (melodies have length) but in changing direction also articulate and organize registral space along a vertical scale (melodies have height). When we imagine melodies as
lines, we imagine their notes as series of points within this two-dimensional mental space
and then connect the dots, as it were, allowing us to picture them as linear contours. Of
course, these melodic lines are never straight: like the two melodies in Ex. 1.1,all melodies outline a series of curves, which meld together to form the larger melodic arc. These
intuitions translate into a number of related metaphors: in addition to those of line and
contour, we often imagine melodies as arcs and curves, ascents and descents, inclines,
grades and slopes, profiles and silhouettes.
In species counterpoint, musicians tend to value pieces in which the two voices trace
distinctive melodic contours and thus assume independent. melodic identities. In Ex. 1.1,
for instance, the upper voice begins with a large leap from D to A, then descends by step
to a low C, leaping again to E before cadencing on D. The lower voice, in contrast, begins
with a brief departure and return to D, continues with disjunct motion first to F and then
to A, and concludes with a long descent from A down to D - it changes direction more
often. Both melodies, in other words, follow different melodic routes to their destinations
and outline different melodic contours: for the most part, the two melodies leap a t different times, reach their melodic peaks a t different times, etc.
Yet the two voices in Ex. 1.1are also similar in a number of crucial respects: we often
exaggerate the extent to which the two voices in a first-species composition are heard as
independent melodies. The upper voice, for instance, begins with a long descent from A,
which is the highest note, or apex, in its melodic curve. As the brackets in Ex. 1.3a suggest,
this melodic descent recurs toward the end of the piece, only now an octave in the lower
voice, where A once again occurs a t the crest of the melodic contour.
.
Ex. 1 . 3 ~
rrnelodic descent from A ...
L m e l o d i c descent from A ...
4
Chapter 1
There is, in other words, a subtle melodic echo in this piece, where the opening melodic
gesture i n the upper voice becomes the closing gesture in the lower voice: while both
voices trace distinct melodic contours, their melodies are cut from the same bolt of musical cloth, and even share melodic material. As the brackets in Ex. 1.3b suggest, we can
even h e a r the jagged ascent in the lower voice between D in ms. 4 to A in ms. 7 as a more
extended and elaborate version of the leapGfromD to A that opens the piece the upper
voice.
Ex. 1.3b
The lower voice takes the opening perfect 5th from D to A in the upper voice and subdivides it, inserting the F in ms. 6 between the D in ms. 4 and the A in ms. 7. It then elaborates the minor 3rd between D in ms. 4 and F in ms. 6 with an upper neighbor G in ms.
5: G, that is, lies "next door" -or rather, "right above" - F in the next measure.
In addition to linear metaphors, we often use stream metaphors to express other intuitions we have about melodies: in species counterpoint, melodies flow toward their destinations -the last note, or final. In contrast to linear metaphors, which encourage us to
view melodies out of time as linear contours that exist all at once at a given moment,
stream metaphors communicate a sense of fluid, ongoing motion toward a goal. In the
course of this advance toward the final, however, melodic motion is never even or continuous, but rather ebbs and flows, like the tide: melodies swirl into eddies at some moments
- pools in which melodies seems to lose momentum or even stagnate - or surge forward
at others. What distinguishes stream metaphors from linear images is the element of motion: stream metaphors are active, dynamic, and temporal, while linear metaphors tend to
be more static and atemporal. Stream metaphors, in other words, stress the sense of motion toward a goal, one of the main musical concerns - if not the main concern - of this
manual. It is not a matter of preferring stream metaphors to linear ones, however, because the two image reservoirs overlap: in the sense we have that melodies define courses
or paths, ascend and descend, or move to the right, stream metaphors subsume linear
ones. While we can imagine lines without motion, one cannot factor out or subtract the
linear element from stream metaphors, and for that reason we will want and need to use
them both.
In general, melodies in species counterpoint are composed of smaller arcs, or gestures,
each of which - in a successful piece - contributes to the overall flow of the tune toward
the cadence. Though somewhat self-contained, melodic gestures are pliable and ofken appear not t o have precise boundaries. Ex. 1.4 thus adds slurs to the score to indicate the
approximate extensions of melodic gestures in both the upper and lower voices:
Ex. 1.4
1
2
3
4
6
6
7
8
9
10
11
Ex. 1.4 is a somewhat crude parsing of these two melodies into gestures, but it gets one
important idea across: gestures tend to conform to changes in melodic contour. Hence the
first four notes in the lower voice meld together to form a gesture, for three basic reasons:
(1)because the gesture begins and ends on the same note, (2) because the last three
notes of the gesture, F-E-D, fill in the initial leap from D in ms. 1to F in ms. 2, and (3)
because a large leap - the perfect 4th from D in ms. 4 to G in ms. 5 - and change in
direction combine to separate the melodic gesture in the opening four measures from the
subsequent melodic continuation as a discrete melodic segment. Likewise, the upper-voice
descent from A in ms. 2 to C in ms. 7 fills in (and then exceeds) the initial leap from D in
ms. 1to A in ms. 2. Here the leap from C in ms. 7 to E in ms. 8 separates the relatively
long opening gesture from the final descent into the cadence. At the same time, however,
the two gestures connect together: the first one flows into the second. In a sense the two
gestures overlap: C in ms. 7 serves as the last note of the opening gesture as well as the
first note of the closing gesture.
Note, too, that both voices udfurl in two more-or-less large gestures, which, however,
do not coincide; in gestural terms, the two melodies are not altogether "in sync" with one
another. A good deal of musical interest in these compositions arises from the subtle interaction between voices, whose component melodic gestures are sometimes in, sometimes out of phase. In this particular piece, the sharing of melodic material between the
two voices - the fact that the final descent from A to D in the lower voice echoes the initial descent from A to D in the upper voice, as mentioned above - contributes to and
complicates this sense of interaction. Melodic interaction, that is, involves more than the
mere coordination of gestures. Here there are two independent melodies that nonetheless
seem to influence one another.
In each voice in Ex. 1.4, the melodic gestures combine to delineate a more comprehensive melodic contour that converges on the final: the final represents a confluence of the
component melodic gestures in each voice. In the upper voice, for instance, the leap from
D in ms. 1to A in ms. 2 opens up a gap that the subsequent melodic continuation fills in.
In its melodic descent from A in ms. 2, the upper voice continues past D to C in ms. 7,
extending the melodic ambitus a wholetone downward. At precisely this moment, however, the upper voice changes direction, leaping back up to E in ms. 8. As it then fills in
the gap, it converges on the final, moving back down through D in ms. 9 to C# in ms. 10,
which, because of the accidental, exerts an almost magnetic attraction (to draw on another set of metaphors ... ) toward the final D in ms. 11.Because the second gap is smaller
than the first one, the entire melodic voice seems to narrow in on the final: the final in
this sense is forehearable; we can sense its imminent arrival well in advance of the actual
event. It thus forms a goal, a terminus toward which the entire melody aims. D - the
note that fills the gap between C and E - is, of course, the final, or rather, is identical to
the final in pitch, but it comes "too soon," two measures before the end. We know that its
6
Chapter 1
arrival is premature because it occurs above an F -rather than a D -in the lower voice.
Realizing that, the upper-voice D seems to press down on F in ms. 9, forcing the lower
voice to descend through E in ms. 10 to the final D in ms. 11.D in ms. 9 forms the moment
of greatest tension - the climax - in the upper voice, because it fills the gap between C
and E, thus "resolving" the leap, but also because we realize that this resolution isn't a
resolution, that we're very near the end but nevertheless still not there. As it guides the
lower voice to the final, the upper voice moves along in parallel 6ths until it reaches C# in
ms. 11, at which point it reverses direction and ascends, via semitone and in contrary
motion with the lower voice, to D in ms. 11.The addition of musica ficta - a C # - in the
penultimate measure lends the arrival of D an air of inevitability. And because in most
cases the final will be identical in pitch to the note that forms the point of melodic departure in the first measure, it will have been in our ears as a potential melodic target from
the very beginning of the piece. In a sense, the cadential arrival on D fulfills an earlier
musical promise: the final forms a conclusion - a moment of completion - rather than a
mere end, a note on which the upper voice just happens to stop.
Ex. 1.5 uses arrows to delineate the confluence of melodic gestures in the upper voice
over the last few measures of the piece. It shows how C, abandoned at the end of the long
opening gesture in ms. 7, curls upward (in the concluding gesture) through C#in ms. 10 to
D in ms. 11;the two melodic gestures flow together into the cadence.
Ex. 1.5
Ex.1.5 thus posits a virtual melodic connection between non-adjacent notes, in this case
between C in ms. 7 and C# in ms. 10. Meanwhile, the concluding gesture splits off from
the opening gesture at C in ms. 7, leaps to E in ms. 8, and then descends v i a stepwise motion through D to C# in ms. 10, where it once again converges with the main melodic current of the upper voice. We can imagine these different gestures as tributaries that follow
their own melodic courses but nevertheless flow into the main melodic current of the
counterpoint.
In both voices, the use of leaps contributes to this sense of forward motion toward the
final. A leap - even a small one - introduces tension into a melody, which manages (or
resolves) that tension via stepwise motion in the opposite direction. In other words, a leap
encourages us to imagine a certain musical continuation, a more or less definite musical
fbture into which the melody will then move. In language used in the last paragraph, a
leap opens up a g a p which the listener expects the counterpoint to then fill.' This dynamic (complementary) relationship between leaps and stepwise motion creates a fluctuation in the level of tension throughout the counterpoint. This is an important point: the
'We owe these terms (and their conceptual elaboration) to Leonard B. Meyer, who first ,introduced
them in Emotion and Meaning in Music (Chicago, 1956). His most extended discussion of gap/ fill occurs in Explaining Music: Essays and Explorations (Chicago, 1973).
7
level of tension in a species composition is never constant, but rather varies throughout
the piece, the point of greatest tension being the climax. The terms climax and apex refer
t o two different melodic phenomena often confused for one another: apex designates the
highest note in a counterpoint (or one of its component melodic gestures), while climax
designates the moment of greatest tension. While the apex and climax can and sometimes
do occur simultaneously, they more ofZen occur a t different points in a piece. The apex
(highest note) of the upper voice in Ex. 1.5, for instance, is the A in ms. 2, while the climax
(moment of greatest tension) comes about as the result of the leap from C to E in ms. 7
and 8. As described above, this leap articulates (gives voice) to two separate melodic
streams (see the arrows between staves) that give the counterpoint the feeling of zeroingin on the final and contributes to a clear sense of forward momentum (or tension) that
does not ease up (or resolve) until C# in ms. 10 ascends to the final D in ms. 11. Note,
moreover, that the apex of the lower voice occurs on A in ms. 7, at the same moment the
upper voice - coincidentally - reaches its lowest note, C. The apex, in other words, occurs a t different moments in the upper and lower voices, another factor that contributes
to the complicated melodic interactions between them. In general, an apex can occur almost anywhere in the first two-thirds of a piece, whereas the climax normally occurs
somewhere in the last third of a piece, at the moment when the final motion into the cadence begins to seem inevitable, a point that most often occurs within three or so measures of the end.
Another important (and for the moment concluding) intuition we have about the
music in Ex. 1.1is that the two melodic voices are harmonious: the two melodies are in
agreement, forming consonances between them. Ex. 1.6includes numerical labels for these
consonances between staves:
Ex. 1.6
A
e
r
r
-
It is clear from these numbers that the piece begins and ends on perfect consonances (the
perfect 8ve), but that in between the two voices form imperfect consonances (major and
minor 3rds [or loths] and 6 t h ~ )In
. the historical tradition from which this music derives,
imperfect consonances were heard to be sweeter, more euphonious, than perfect consonances, which in comparison were thought to sound austere. In this particular piece, imperfect consonances far outnumber perfect consonances, which is one of the reasons the
music sounds so rich and sonorous.
1.2 Performance
In the last section, we discussed some of the basic characteristics of species counterpoint and considered one particular piece in some detail. Up to now, our comments on the
performance of this music have been somewhat informal, amounting to little more than
Chapter 1
8
the common-sense advice to draw on one's own intuitions about how best t o sing the
music. We will now address issues of performance more directly and develop more specific ideas as to what a musical performance of this piece would entail. We want to stress
that even though species counterpoint is primarily used for pedagogical purposes as a
controlled. musical environment within which to compose melodies and conceptualize
melodic behavior, these pieces are real musical compositions, each with its own integrity,
and as such deserve caring performances - all music deserves to be well-performed.
More than that, however, the principles that inform the performance of species compositions can be applied to all sorts of other music and thus have a broader, more global relevance: we can draw on these same principles to put together a more musical performance
of almost any piece of music.
Vocal Production
Historically, species counterpoint was modeled on Renaissance vocal music, and therefore sounds best when sung; the performance aesthetic we are aiming at is a vocal one. In
singing this music, use a light but focused tone, no vibrato. Men with lower voices will
need to sing, comfortably, in a middle or higher register, with a range that extends from
an octave or so below middle C up to F or G above middle C . Since species counterpoint
involves a concern with both the contours of individual melodies as well as their harmonious coordination, our goal in performance will be to blend the sound of the two voices,
something that becomes difficult when baritones and basses, in particular, sing in their
lower registers, an octave below the tenors. As you sing your part, you need to be aware
of the other part as well, so that you can hear the relation between voices: never sing so
loudly that you can't hear the other singer (or singers). In order to get a good aural sense
of the composition as a whole, it is important to rehearse the piece, and it is especially
valuable to trade parts, which often helps us in learning to hear the piece as a blend of
melodies as we continue to sing one of them.
If possible, sing the melodies in a single breath - easier to do in first species than in
the others - and maintain a smooth, legato connection fi-om one note to the next: there is
a close aesthetic correlation between conjunct (stepwise) melodic motion and legato singing. Singing these melodies legato, moreover, assists us in both creating and conveying
forward motion in this music: a sense of musical flowis easier to realize when we imagine
legato singing as a transfer of momentum from note to note, rather than merely as the
smooth connection between them. Some musicians associate legato with relaxation, with
a serene tensionlessness, but nothing could be farther than the case: legato is one of the
principal means a performer has of creating and sustaining musical tension, in even the
softest and slowest of music.
Gesture
Ex. 1.7 (on p. 9 below) translates the slurs in Ex. 1.4 above into crescendos and decrescendos; again, these markings represent a crude parsing of these melodies into gestures, for in actual musical practice, melodic gestures blend and meld into one another. A
basic (or a t least traditional) principle of performance is that one crescendos when ascending and decrescendos when descending. If we were to sing Ex. 1.7 like that, however,
our performance would sound somewhat mechanical if not downright unnatural, the
result of a n overly-literal application of a crude and unnuanced generalization about performance - a simplistic, one-to-one correlation between volume and melodic direction. A
,
Ex. 1.7
more sophisticated and musical approach would be to use vocal intensity - which involves much more than volume - to delineate and underline melodic gestures within the
melodies: the idea, in other words, would be to use subtle, almost imperceptible changes
of volume and intensity to perform gestures. In general, there will be a gradual increase
in intensity toward the middle of a gesture, at which point intensity will decrease and subside until the next gesture begins. In a musical (interesting) performance, the gradual
wax and wane in vocal intensity will correspond to subtle increases and decreases in volume: the hairpins in Ex. 1.7, then, are meant to bring out this general correspondence.
Sing the two melodies again with this principle in mind: crescendo (as it were) into each
gesture, then decrescendo out of it; you will find the results more satiswng. Of course,
questions of where, precisely, to locate these hairpins and how best to realize them are
open to discussion; it is these sorts of discussions that we are hoping to encourage.
The principle of increasing and decreasing intensities can also be applied to how we
control the motion of these melodies through time, to how we pace the music. Once again,
as we move into a gesture, there will be a gradual, almost imperceptible increase in tempo, and a similar broadening in pace when moving out of a gesture, all corresponding to
the changing musical intensities of the music and done in a manner that enhances a sense
of seamlessness. We need to emphasize that these changes in pacing need to be subtle in
the extreme: when done well, the listener won't be consciously aware changes are taking
place. The effect, rather, will be one of a gradual compression and decompression - contraction and expansion, inhalation and exhalation - of musical time within a single tempo. These subtle manipulations of musical time should be used with great care; however
essential to realizing the ebb and flow of this music, these sorts of nuances can be easily
overdone, resulting in exaggerated, unmusical (obvious and insensitive) performances.
In a musical performance, singers will intuit how melodies are composed of smaller
gestures and communicate that awareness to one another, and to the listener: simply put,
a musical performance is one in which the performer delineates the gestures that combine to form melodies. In this case, a sensitive performance of the music would allow the
concluding gesture in the lower voice to emerge out from under the shadow of the upper
voice as it nears the cadence, bringing out its melodic echo of the long descent from A in
the opening measures of the upper voice (as discussed in connection with Ex. 1.2a). As the
opening gesture in the upper voice subsides, following the decrescendo above the staff in
Ex. 1.7, the lower voice grows in presence, following the crescendo below the staff' in Ex.
1.7, emerging into our conscious awareness to the point that it seems to guide the upper
voice into the cadence.
Chapter P
Solfege
Solfege refers t o the practice of using syllables to vocalize pitches, but also as mnemonic devices for singing melodic intervals. It is used to assist in the aural transmission of
music and exists in one form or another in most musical cultures. The form that has come
down to us is now a millennium old: it has been often attributed to Guido d'hezzo, a reknowned medieval musician who died sometime after 1033. Guido used solfege as a pedagogical device to teach arrangements of wholetones and semitones in the absence of musical notation; it has undergone modification and been used for different musical purposes
ever since. We know it from The Sound of Music: "do, a deer, a female deer; re, a drop of
golden sun ...."2 One derivation of the word solfege, which embraces multiple meanings,
traces it from an amalgam of sol ("a needle pulling thread") and fa ("a long, long way to
go").
Over the last two centuries, the two most common methods of solfege have been fixed
do on the one hand and moveable do on the other. In fixed do solfege, syllables are used
to designate specific letter-named pitches: do = C, re = D, mi = E, and so on. In moveable do
solfege, syllables designate the order position of each pitch in a scale: do =the first note of
the scale, re = the second note of the scale, mi = the third note, and so on. Both methods
have their uses, but for singing species counterpoint, fixed do is more appropriate; in
fixed do solfege, the interval between mi and sol, for instance, will always be a minor 31-4
no matter what mode (see below) a piece is in; that would not be the case were we to use
moveable do.
We will use a form of solfege, moreover, that can be described as funtional solfege, in
which si is used for "raised" pitches, or in the case of species counterpoint, leading tones.
C# in the penultimate measure of Ex. 1.1,for instance, is a raised leading tone: Ex. 1.1,
that is, uses an accidental to raise a CLI to a C#, which creates an artificial semitone - a
semitone where a wholetone would otherwise have occurred - between C# and D; as discussed above, the semitone creates an almost gravitational attraction between C# and D
and thus intensifies the sense of forward motion into the final. Because si forms a semitone in relation to do in the unaltered diatonic collection, we will label C# in Ex. 1.1in
functional terms as a si in relation to re.
1.3 Basic Concepts
Cantus firmus
In composing species counterpoint, the basic method is to add a melodic part, or counterpoint, to a cantus firmus, or "fixed voice" - cantus for short.3 A cantus firmus is some
pre-existent tune used as the basis for a multivoice piece: in species counterpoint, the
same cantus firmus often forms the basis for numerous different compositions. It is a rem-
,
The introduction of solfege into The Sound of Music in fact constitutes an (unintended)historical re
enactment of its invention. Guido f0und.a hymn, Ut queant laxis, in which each phrase begins on
the next note in the scale, which he named with the Latin syllable sung on that note: he thus
named C ut (which later became do), D re, E mi, and so on. Which is precisely what Rodgers & Hammerstein do: the first line of the song begins on C, the second on D, etc. Rodgers & Hammerstein also introduce solfege for the same reason Guido did: Maria uses it in teaching the children to sing.
Note the two different uses of counterpoint in this sentence, in which the term refers t o a particular
melodic voice on the one hand, but also to an entire genre of music on the other.
11
nant of an ancient historical practice in which plainchant melodies were used as the bases
for liturgical music (sacred motets, mass movements, etc.), a compositional procedure that
had all but died out by the time J.J. Fux codified counterpoint instruction in terms of five
species (or metrical states) in Gradus ad Parnassum (1725). Most of the cantus firmi used
in species counterpoint, however, have been composed for the express purpose of writing
counterpoint above and below them, and do not derive from liturgical melodies.
Species counterpoint has come down to us with its own repertoire of cantus firmi, a
few of which, in certain pedagogical contexts, have become almost famous: the cantus firmus on the C clef in Ex. 1.1is the most famous of them all, one that Fux used in Gradus
ad Parnassum, and which has circulated ever since in species-counterpoint manuals and
in the context of composition instruction. In species counterpoint, a cantus is firrnus in
two senses: (1)it operates as a -concretemelodic arch around which the entire composition
is constructed, and with which the counterpoint agrees and conforms, and (2) it is carved
in stone, as it were, and unalterable: when composing counterpoint, no changes are to be
made to the cantus firrnus. A traditional species-counterpoint cantus firmus, moreover,
comprises nothing but whole notes, one per measure, and most ofhen ranges between 9
and 13 notes in length. It begins and ends on the same pitch, the mode final, which it
invariably approaches from above, descending from 3 to i, where the carets designate
degrees of the modal scale (or mode degrees) and the final is numbered 2. Hence the
cantus firmus in Ex. 1.1concludes with stepwise motion from E in ms. 10 to the final D in
ms. 11.
Mode
In a species composition, then, the final is the last note in the cantus firmus, but it is
also the most important. It determines a locus position within the music in relation to
which we hear other notes and calculate the distances between them. It allows the cantus
firmus to define a particular melodic environment, a mental grid of wholetones and
semitones against which we hear the music. In modal music, the crucial factor is how
semitones are distributed within the octave above the final. Without going into detail, Ex.
1.8 gives the modal octaves for the six modes that begin on notes of the natural hexachord
(C-D-E-F-G-A) and identifies the location of wholetones (Ts) and semitories (Ss):
Ex. 1.8
A
.. -
A
* , ,
w
-
-
S
I
T
T
s
T
--
"
T
T
T
T
S
u
u
a
s
T
--
A
D
m
-CcC#"
-
-
m.
A
r
,
T
T
T
s
T
T
S
T
T
S
T
T
S
T
F
T
T
G
Chapter 1
12
Each mode thus articulates its own distinctive pattern of wholetones and semitones. In
the D mode, for instance, the two semitones fall - as in no other mode - between P and 3
and S and 9. In the C mode, in contrast, the semitones fall between 3 and 2 and ? and $, also
a unique pattern among the six modes.
This, then, is the reason for using fixed do solfege: to preserve the unique scalar identities of the six modes. Under fixed do, semitones will always occur between mi and fa and
between si and do, regardless of the mode. That would not be the case were we to use
moveable do.
Musical Goals
A good species composition will meet three basic criteria: (1) the counterpoint and
cantus firmus will agree with one another, allowing listeners to concentrate on the
melodic profile of each voice, (2) both voices will flow into their cadential goals, and
(3) each counterpoint will have its own identifiable melodic contour, distinct from but still .consistent with - that of the cantus firrnus.
Over a number of centuries, musicians have devised a set of integrated rules and procedures designed to meet these objectives and ensure that their melodies will be musical. In
a moment, we will turn to those rules and procedures, but in order for that material to be
comprehensible, we will need to familiarize ourselves with some additional concepts and
information: information about cadences on the one hand, and concepts about relations
between voices on the other.
Consonance and Dissonance
Consonance and dissonance are among the most difficult and problematic concepts in
music: despite our continual reliance on them, we are unable, even after centuries of
hard labor, to give them precise definitions. Ideas about consonance and dissonance, moreover have changed a great deal over historical time: medieval and renaissance musicians,
for instance, believed that consonances (from the Latin consonare, to sound together) were
euphonious, sweet sounding, and that dissonances were non-euphonious - consonances,
that is, agree, while dissonances disagree. This is, of course, an assertion rather than an
explanation - it does not tell us what it means for consonances to agree - but we will
nevertheless use it as a provisional point of departure.
For now, we will define consonances as perfect unisons, Sths, and 8vek, plus (imperfect) major and minor 3rds and Gths. We will then define dissonances negatively, as nonconsonances: all major and minor 2nds and 7ths, plus all diminished and augmented intervals; the perfect 4th has an in-between status, but for the purposes of species counterpoint will be listed among the dissonances. There are of course different degrees of consonance and dissonance, but now, before we've become familiar with the music, is not the
time to consider them; we will return to the topic now and then, as it arises from other
musical considerations.
Melodic Range (Ambitus) and Distance between Voices
The melodic range (or ambitus) of a counterpoint should be relatively modest. It will
rarely exceed a n octave and should be roughly proportionate to that of the cantus firmus:
in Ex. 1.1,for instance, the range of the counterpoint is a major 6th, while the range of
the cantus firmus is a perfect 5th - the two ranges are proportionate to one another. Nor
should the distance between voices be too great. While there's no reason t o set a definite
limit on the acceptable distance between voices, separations of between a 3rd and a 13th
(an octave and a 6th) are common. When the voices wander too far apart, it becomes
difficult for us to blend the voices together.
Simple and Compound Intervals
For our purposes, compound intervals will be intervals of a 10th or larger, while simple intervals will be those of a 10th or smaller. (This is not the conventional definition of
the term, which gives the octave as the cutoff point.) As a convenience, we will label compound intervals according to their simple equivalents. We will thus label the interval from
C up to A in ms. 7 of Ex. 1.13b on p. 18 below as a (major) 6th, even though, in real terms,
it forms a 13th: most of us have would have to do a little arithmetic to figure out that an
octave and a 6th add up to a 13th.
Four Varieties of Relative Motion
There are four possible varieties of relative motion between the cantus firmus and the
counterpoint: parallel, similar, oblique, and contrary. Relative motion refers to the direction of motion in the counterpoint visd-vis melodic direction in the cantus firmus.
Parallel motion occurs when both voices move in the same direction and maintain
the same size intervals between them. As the first bracket in Ex. 1.9a indicates, the
cantus firmus and counterpoint move together in parallel 1 0 t h ~
from ms. 2 through
ms. 4: though some of these lOths are major, and others minor, the two voices are
still considered to be moving in parallel motion.
Ex. 1 . 9 ~parallel
:
motion (p)
I
p
I
L pJ
I
n
-
Similar motion occurs when both voices move in the same direction but the intervals between them change. In the first two measures of Ex. 1.9b, for instance, the
cantus firmus ascends a minor 3rd from D to F while the counterpoint ascends a
perfect 5th from D to A, forming first a perfect 8ve, then a major 10th: both voices
ascend, but the interval between them changes from ms. 1to ms. 2.
Chapter 1
Ex. 1.9b: similar motion (s)
Oblique motion occurs, in first species, whenever there's a tie in the counterpoint.
~ ~ cantus firmus ascends
Because of the tie in the first two measures of Ex. 1 . 9 the
from D to F against a sustained D in the counterpoint, creating an oblique succes- .
sion from a perfect 8ve to a major 6th.
Ex. 1 . 9 ~oblique
:
motion (0)
Contrary motion occurs when both voices move in opposite directions. In the last
two measures of Ex. 1.9d, the cantus firmus descends from E to D while the counterpoint moves in contrary motion from C# to D. Contrary motion allows for maximal contrast between voices in terms of melodic direction, and for that reason is
used (as it is here) to form cadences.
Ex. 1.9d: contrary motion (c)
All four varieties of relative motion are good and useful, and for each one there will
be occasions when none of the other varieties will do. To some extent, this classification is
more discursive than musical: it allows us to discuss relations between voices with greater
ease. It is often noted that the four forms of relative motion create a continuum, in the
order given above, ranging from least to most contrast in melodic direction between voices.
It is not true, however, that independence of voices takes precedence over other musical
considerations, or that contrary motion is for that reason preferable to the other three
varieties of relative motion, which would be wrong. In composing first-species counter-
15
point, there will be times when contrary motion - required at cadences, and indispensable a t other times - will be worse than useless. As a general rule, the most useful form
of relative motion in first species is in fact parallel motion, which often accounts for half
or more of the successions from one harmonic interval to the next in an average firstspecies composition, as it does in Ex. 1.9a, where 5 of 10 total successions are in parallel
motion.
A brief digression: Ex. 1 . 9 ~
is the same as Ex. 1 . 9 except
~
for the tie in the first two
measures. The presence of the tie, however, makes quite a difference. Instead of leaping a
begins with a more modest leap of a
perfect 5th from D to A, the counterpoint in Ex. 1 . 9 ~
perfect 4th from D to G, which narrows the melodic ambitus. It also ruins the melodic
echo between the opening melodic descent from A in the counterpoint of Ex. 1 . 9 ~and the
final melodic descent from A in the cantus firmus. Another melodic echo, however, comes
about as a result of the tie, as if in compensation: the initial D-G-F gesture in the counterpoint of Ex. 1 . 9 ~recurs moments later in the cantus firmus. But that gesture is not as
self-contained as the melodic descent from A in Ex. 1.9a, nor is the gesture used - as the
melodic descent from A is in Ex. 1 . 9 ~
- to round off the cantus firmus: Ex. 1.9a, that is,
transforms its opening gesture into a closing one. And because of the tie, it takes Ex. 1 . 9 ~
longer to gain melodic momentum, to generate forward motion toward the cadence.
is less compelling than the one in
In some respects, then, the counterpoint in Ex. 1 . 9 ~
Ex. 1.9a. Yet the conclusion to be drawn here is not so much that the counterpoint in Ex.
1.9~
is better than the one in Ex. 1.9c, but first, that even an innocuous change in melodic
contour can have rather dramatic consequences, and second, that the concepts and principles of species counterpoint allow us to make these sorts of comparisons and to discuss
them with real precision.
Clausula formalis
All species counterpoints, without exception, end with a clausula formalis, or formal
close, which serves as musical punctuation to bring the composition to a strong conclusion. Use of the clausula formalis is not confined to species counterpoint: it forms the primary gesture of musical closure in an enormous repertoire of music dating from the 14th
through the 17th centuries; species counterpoint borrowed the clausula formalis from
that repertoire.
In the clausula formalis, the counterpoint cadences on a perfect unison or a perfect
8ve above or below the cantus firmus; the approach to the cadential consonance, moreover, must be stepwise in both voices, via contrary motion, and one of the voices must
move by semitone - these are absolute requirements. In Ex. 1.10, the counterpoint thus
moves from C# to D in contrary motion to the cantus firmus, which moves from E to D a semitone in the counterpoint against a wholetone in the cantus firmus.
Ex. 1.10
16
Chapter 1
In addition to the contrast in melodic direction between voices, the rules for cadence formation also ensure maximal contrast between the penultimate imperfect consonance (the
major 6th E-C#) and the perfect consonance (the perfect 8ve D-D) in the last measure.
In three of the six modes - the D mode, G mode, and A mode - the rule requiring
semitone motion in one voice also requires us to raise the next-to-last note in the counterpoint, converting the mode degree below the final into a subsemitonum modi, or leading
tone. In these same three modes, this necessitates use of an accidental: as a consequence
of the accidental in Ex. 1.10, C# in ms. 10 leans toward the cadential D in ms. 11;the chromaticism intensifies the melodic surge to the final.
In all but one mode - or rather, in all the modes that have one - the leading tone
will occur in the counterpoint, the added voice. In the other mode, the E mode, there is no
subsemitonum modi. As Ex. l.lla demonstrates, raising D to D# in the E mode would
create a dissonance (an augmented 6th) with the cantus firmus. Because harmonic intervals are restricted to consonances in first species, the lower leading tone in the E mode is
left unaltered, as in Ex. 1.llb. Here the two voices form a phrygian cadence, so called
because the cantus firmus (rather than the counterpoint) moves via semitone to the mode
final: in renaissance music theory, the E mode was sometimes identified (using an arcane
Greek name) as the phrygian mode.
Ex. 1-lla
Ex. 1 - l l b
In the A mode, raising Gh to G# causes a melodic dissonance (an augmented 2nd) when the
counterpoint approaches G# from F, as it does in Ex. 1.12a. In order to eliminate the
augmented 2nd - but also to accelerate melodic motion into the final - F must be raised
to F# when approaching G# from below, as in Ex. 1.12b.
Ex. 1.12~
1
2
3
4
5
6
7
8 9
r A 2 1 1
1
Ex. 1.12b
Raising F to F#,incidentally, has consequences for our method of solfege. Just as the leading note G# in ms. 8 is no longer a simple sol in relation to la - the final A - in ms. 9, F#
in ms. 7 is no longer a simple fa. And just as the leading note C# forms a si in relation to re
in the D mode, F# and G# form a la and si in relation to the final la - A - in the A mode.
Hence we would sing the counterpoint in Ex. 1.12b as la-si-do-si-la-sol-la-si-la,where the
italics designate the altered 8 and 7.
N.B. In species counterpoint, accidentals are used in the formation of cadences, but
at no other times, and for no other reasons.
1.4 Rules
There are two basic categories of rules in species counterpoint: one that governs vertical relations (relations of consonance and dissonance) between voices, and another that
governs the melodic behavior of individual horizontal voices. In species counterpoint, the
horizontal/melodic dimension always takes precedence. The rules governing verticallharmonic relations between voices, in contrast, are meant to ensure that the two horizontal
voices don't interfere with one another, and that our attention won't be distracted from
following the melodic progress of individual voices.
Rules governing relations between voices are more stringent than those governing
melodic behavior. For that reason, we will describe the former as well-formedness rules,
the latter as preference rules.4 If our compositions are to be recognizable to other listeners
as species counterpoint, we must follow these well-formedness rules - there's no choice
in the matter. When we depart from these rules, other listeners will either regard those
The distinction between well-formedness and preference rules derives from Fred Lehrdahl b d Ray
JackendoffsA Generative Theory of Tonal Music (Cambridge,'Mass., 1985), where it forms the basis
of an ambitious, rule-driven model of musical cognition.
Chapter 1
18
departures as mechanical errors, or (if the errors are numerous enough) will no longer
hear the music as species counterpoint. Which doesn't necessarily mean, of course, that
the music won't sound good - to us, if not to other listeners - but rather that we will not
be able t o use the basic principles of species counterpoint to describe, critique, or improve
our melodies, or the melodies of others.
Most of the following rules represent more precise formulations of ideas familiar to us
from our discussions of Ex. 1.1 above.
Rules Governing Relations between Voices
1. In first-species counterpoint, only consonances occur, and those are: the perfect unison,
5th, and 8ve, plus (imperfect) major and minor 3rds and 6 t h ~ .
2. The counterpoint must begin on a perfect unison, perfect 5th, or perfect 8ve above the
cantus firmus, or - when the counterpoint begins below the cantus firmus - a perfect
unison or perfect 8ve below.
This means that there are fewer options for beginning a counterpoint below the
cantus firmus than beginning above. Once again, the reason has to do with the
modal nature of the music. In addition to the final, a mode attaches prime importance to where (in all but one case) the perfect 5th between i and B falls within its
particular melodic environment of tones and semitones. When the cantus firmus
begins on D and the counterpoint begins above on A, as it does in Ex. 1.13a, the
perfect 5th between them conforms to the D mode and affirms the modal identity
of the cantus firmus, which we will expect to close on D.
Ex. 1 . 1 3 ~
If, however, the counterpoint were to begin on a G below, as it does in Ex. 1.13b,
the perfect 5th between G and D would conform to the G mode. It would surprise
us if both the cantus firmus and the counterpoint were then to close on D rather
Ex. 1.13b
than G. In other words, beginning on the perfect 5th below causes confision about
the mode.
Now that the situation has arisen, note that the two voices in Ex. 1 . 1 3 ~
cross in ms.
5, where the cantus firmus leaps over the counterpoint to G and remains above
the counterpoint straight through to the cadence. A voice crossing occurs whenever one voice leaps over or under the other, and while voice crossings are more
common and easier to manage in second species, they're perfectly acceptable in
first species as well, though there's no particular virtue to them, either.
3. The counterpoint must close on either a perfect unison or a perfect 8ve above or below
the cantus firmus, and must form with the cantus firmus a clausula formalis. For a complete description of the clausula formalis, see pp. 15-17 above.
4. In between the first and last measures, avoid perfect unisons and 8ves, which, amid the
spare textures of first-species counterpoint, sound austere.
There is, however, an important exception to this rule, though few other manuals
on species counterpoint make allowance for it, and that is the voice-exchange octave. In Ex. 1.14, a voice-exchange octave occurs above the E in the cantus firmus
a t ms. 3. It occurs within a three-measure gesture in which the outer voices move
by step in contrary motion, forming the consonances 6-8-10 or - depending on
the melodic direction of the cantus firmus - the consonances 10-8-6. In both
cases, the cantus firmus and counterpoint pass through the octave: there is an inbetweenness to the voice-exchange octave that softens the perfect 8ve and causes it
to sound less austere. In a sense, the cantus firmus and the counterpoint trade (or
exchange) pitch classes: the large X in Ex. 1.14 points out that F in ms. 2 of the
cantus firmus turns up again in ms. 4 of the counterpoint, while D ms. 2 in the
counterpoint turns up again in ms. 4 of the cantus firrnus.
Ex. 1.14
Another voice exchange occurs between ms. 3 and 5 in Ex. 1 . 1 2 ~and Ex. 1.12b on
p. 17.
Voice-exchange octaves occur all over the place in music from the 17th through
the 19th centuries and are sometimes used to control and organize music over long
spans of time. First-species counterpoint presents us with the perfect musical conditions in which to become familiar with them.
.
5a. Parallel perfect consonances never occur in species counterpoint, for reasons to be
discussed later.
Chapter 1
20
5b. Within the aesthetic that governs species counterpoint, perfect consonances are
thought to sound austere, and for that reason require special treatment. In addition to the
prohibition on parallel perfect consonances (see 5a above), avoid moving to perfect 5ths
and 8ves in similar motion. Approaching perfect 5ths and 8ves in similar motion "exposes"
them and makes them sound even more severe than usual; perfect consonances approached in similar motion are known as direct 5ths and Sves.
In Ex. 1.15, motion in the same direction in both the cantus firmus and counterpoint result in a direct 5th above A in ms. 7: the cantus firmus leaps a major 3rd
from F up to A, while the counterpoint moves up by step from D to E, moving, that
is, in similar motion.
Ex. 1.15
t
direct 5th
There is another perfect 5th in E x 1.15, above F in ms. 2, but in this case the perfect 5th is not direct: it is approached, rather, in contrary motion, the cantus firmus
leaping up a minor 3rd from D to F, the counterpoint moving down by step from D
to C.
Perfects 5ths and 8ves, in other words, should be approached in either contrary or
oblique motion.
Because of the limitations on the use perfect 8ves in first species, direct 8ves are
rare in first species, occuring far more often in second and third.
It is not enough, however, for a counterpoint to follow the rules, because the rules
governing relations between voices will never ensure that the tune will be a good one. In
order to help ensure that a counterpoint will be musical, we will introduce a second set of
rules governing melodic behavior. These are preference rules, rules that enter into decisions about what to do in a given context, but which do not in themselves determine a correct solution or prescribe a precise course of action. A preference rule is never a matter of
right or wrong. It requires us, rather, to use our musical intuitions to decide whether a
rule applies and how it applies in a given situation. In most cases, a preference rule will
suggest a number of possible solutions from which to choose, or a more general course of
action; unlike well-formedness rules, there's wiggle to them. Preference rules are not, however, optional, but rather isolate and express some of the core principles of good melodic
construction, and are in force at all times.
21
We can summarize preference rules in very general terms: in species counterpoint,
good melodies (1) have simple contours that flow to the cadence, (2) sound natural
and are comfortable to sing, and (3) are relatively easy to remember.
6. Good melodies are continuous, integrating smaller gestures into larger, more comprehensive melodic contours. For that reason, avoid the repetition of small gestures - or doing anything else - that would contribute to breaking the counterpoint up into discrete
melodic fragments. Good melodies are all of a piece, flowing seemingly in a single motion
from beginning to end. In a sense, most of the other rules governing melodic behavior
represent more specific formulations or refinements of this one basic principle.
Heinrich Schenker, an important writer on species counterpoint, understood the continuousness of good melodies in terms of a melodic "equilibrium" in which all notes receive
the same amount of emphasis. 'We must," he writes, "aim for a complete equilibrium of the
tones in relation to each other, in contrast to the predominance of individual, independent
fi-agment~."~
7. Each melodic voice should have its own distinctive melodic contour. This means, to begin with, that each voice will have a melodic apex -a single note that articulates the high
(or in some cases low) point of the entire melodic arc - but also that the melodic apex in
the counterpoint should not occur a t the same time as that in the cantus firmus.
A commonly proffered rule in first-species counterpoint is that one should avoid
more than three consecutive consonances of the same size; that is, there should
never be more than three 3rds, three Gths, or three 1 0 t h ~
in a row, the idea being
that if thebcantus firmus and counterpoint proceed together in parallel motion for
more than three measures, the two voices will lose their independence. Not a good
rule. To begin with, the rule is arbitrary: why three measures and not four? Nor
does it take the length of the cantus finnus into account: four consecutive parallel
consonances will affect the independence of voices differently in a nine-note cantus firmus than it will in one of thirteen. One of the important lessons to be learned
fkom species counterpoint is that the effect of a given compositional gesture has to
be considered in context and therefore on a case-by-case basis.
It would be difficult, moreover, to imagine someone objecting to four consecutive
parallel consonances on the basis of their sounding bad. In the enormous historical
repertoire of tonal music, there are countless extended passages and even entire
pieces in which two voices remain in parallel. 6ths or lOths for almost the entire
time. Parallel motion is basic to tonal music.
If the point is to ensure the independence of voices, it is better to introduce a positive requirement to that effect, as 7 does, than to place mechanical restrictions on
our compositional options. As a general rule, even a small contrast between the
Heinrich Schenker (1868-19351,Counterpoint, Vol. 1(19101, tr. John Rothgeb (New York, 1987), p.
18. Schenker's massive, two-volume treatise is the most extensive consideration of species' counterpoint ever to appear; in the German original, for instance, he devotes a full 126 pages t o the cantus
firmus alone!
Chapter 1
22
two voices - a well-timed leap, for instance, or the use of contrary motion near
the beginning of the piece - will be more than enough to establish the independence of voices for the duration. And the clausula formalis ensures that our last
impression of a piece will involve the motion of both voices in opposite directions
to the final: the one most crucial melodic gesture in the entire piece serves to reemphasize the independence of voices.
8. Good melodies are primarily stepwise, with a only few leaps added in.
In Ex. 1.1above, there are three leaps in the cantus firmus and two leaps in the
counterpoint, which for a cantus eleven notes long is about average.
9. Leaps create tension, and tension requires resolution. For that reason, follow leaps
larger than a major 3rd with motion - preferably stepwise - in the opposite direction. It
is also a good idea to precede (or prepare) leaps larger than a major 3rd with motion again, preferably stepwise - in the opposite direction.
Preferably, not necessarily, stepwise. 9 is apreference rule: do not interpret it as an
iron rule of law; doing so will make composing much more difficult.
9 is meant to ensure that leaps are integrated into melodies, that leaps, more to
point, are used to join melodic gestures together rather than break the counterpoint into disconnected melodic segments. In the approach to the cadence in Ex.
1.1, 9 ensures that the two main melodic gestures in the counterpoint - the
melodic descent that concludes on C in ms. 7 and the melodic figure that begins on
E in ms. 8 and zeros in on the final - will overlap in register, allowing us the integrate them into the larger melodic contour of the counterpoint. In this case, the
opening leap from D to A is both resolved and filled in, as is the leap fiom C to E
near the end.
9 does not stipulate that a leap has to be filled in immediately, though in most
cases that's what happens, a t least when the leap is a relatively small one, a 3rd or
a 4th. In the cantus firmus of Ex. 1.1, for instance, the leap from D in ms. 4 is resolved (with stepwise motion down to F in ms. 6) but not immediately filled in. We
have to wait until E in ms. 11for to gap to be entirely filled.
10. Avoid dissonant leaps, which rules out major and minor 7ths, diminished Bths, and
augmented 4ths. And avoid melodic contours that outline dissonant intervals.
11.It is preferable to approach the leading note via stepwise motion. In first species, that
means the leading note will almost invariably be approached from above.
12. It is acceptable to use ties in first-species counterpoint to allow for greater composi~ Ex. 1.14) or to evade problems that would othertional ease and freedom (as in Ex. 1 . 9 or
wise arise without them. I t is obvious, however, that a tie will inhibit melodic momentum,
and for that reason never use more than one p e r composition. Without exception, ties
work best a t the very beginning of ,a piece, before the counterpoint has begun to generate
real momentum. A tie toward the end, in contrast, can be disastrous,'bringing melodic
motion to a standstill at the precise moment when it most needs to move forward ti the
final.
-
23
13. No melodic chromaticism. Ex. 1.16 demonstrates how melodic chromaticism can arise
in first-species counterpoint when no other rules appear to have been violated. In this
case, C forms a perfect 5th above F in ms. 9 and then moves a chromatic semitone to C#,
the leading note, which forms a major 6th (the penultimate imperfect consonance) above
E in ms. 10. Melodic chromaticism arises, that is, near the cadence, in conjunction with the
accidental raising the leading note, which is when to be on the listen for it. Unlike most of
the other rules governing melodic behavior, this one is for all intents and purposes a
well-formedness rule.
Ex.1.16
13 affects neither the semitone from B in ms. 8 to C in ms. 9 nor the semitone from
C# in ms. 10 to D in ms. 11,both of which are proper to the mode.
1.5 How to Compose First Species Counterpoint above a Cantus firmus
In this section, we will walk through the process of composing counterpoint above a
cantus firmus. First, some procedural suggestions. It is above all essential to listen to the
counterpoint you're composing as you're composing it: if at all possible, c&npose at the
piano so that' you can hear the two voices together. You should also be singing the entire
time to make sure the counterpoint conforms to the melodic principles discussed above. A
good way to begin composing counterpoint - the best way, in fact - is to memorize the
cantus firmus, singing it over and over until you are comfortable with its twists and turns
and can give a good performance of it; sing the cantus firmus until you know it by heart.
Generally, a good melody will be easy to sing: if you are having a hard time singing a
counterpoint, more often than not it will be because it runs afoul of one or more well-formedness or preference rules. Composing species counterpoint should not be an abstract,
intellectual exercise, but rather a matter of creating real music (which it is!) intended for
actual performance.
Remember: a good counterpoint will comprise a series of gestures that meld together
and flow into the cadence - listen and compose in larger motions rather than individual
notes. If possible, put notes down on paper two, three, and even four at a time: compose
in multinote segments, in gestures. The more one composes, the easier this becomes to
do; although the rules will seem cumbersome and constraining a t first, the melodic principles embodied in those rules will soon become second nature. You may eventually be
able to improvise a counterpoint to a cantus firmus, spontaneously, all in a single breath.
In this tutorial, we will compose a first-species counterpoint above the G mode cantus
firmus in Ex. 1 . 1 7 ~ .
Chapter 1
Ex. 1 . 1 7 ~
1
2
3
4
5
6
7
8
9
10
11
12
13
A
We can hear this cantus firmus as an combination of three large gestures: (1)a stepwise
melodic ascent from G up to E, (2) a series of three consecutive leaps connecting E with
D, and (3) a stepwise melodic descent from D down to G, the final. Of these three
gestures, the second one is the most complex. After the opening gesture rises to E in ms. 6
- a note we will come to remember as the apex of the entire tune - the second gesture
takes over and descends to A in two stages, leaping a major 3rd from E to C and then a
minor 3rd h m C to A. In other words, it subdivides the perfect 5th between E and A into
two smaller intervals, and major 3rd and a minor 3rd, and for that reason we will refer to
the larger interval as a subdivided leap. Even though the perfect 5th (in this case) is a
composite of two smaller intervals, it has the cumulative effect of a single large leap, and
because a perfect 5th is larger than a major 3rd, it must continue with motion in the opposite direction. The cantus firmus thus ascends from A in ms. 8 to D in ms. 9, resolving
the subdivided leap. It does not, however, resolve the subdivided leap with stepwise motion in the opposite direction, nor does it have to: the pertinent rule (9 on p. 22) states that
a leap larger than a major 3rd must be followed with motion in the opposite direction,
preferably - but not necessarily - by step. As a perfect 4th, the leap from A in ms. 8 to
D in ms. 9 must also continue in the opposite direction: the stepwise melodic descent from
D in ms. 9 to G in ms. 13 thus fills in two gaps, the perfect 4th that opens up between A in
ms. 8 and D in ms. 9, but also the earlier subdivided perfect 5th between E in ms. 6 and A
in ms. 8. Because the cantus firmus leaps from A in ms. 8 to D in ms. 9, it resolves the
subdivided leap between E in ms. 6 and A in ms. 8 without filling it in, an obligation which
then falls to the melodic descent from D in ms. 9: in the process of descending to the final
G i n ms. 13, that: stepwise descent fills in the earlier gap. In a sense, the subdivided leap
energizes the cantus firmus: most of the cantus firrnus from ms. 8 on can be heard as a response to the subdivided perfect 5th, as an effort to resolve the tensions it introduces into
the tune.
So much for the cantus firmus; now we need to compose a counterpoint to go above it.
When composing species counterpoint, the best approach is to come up with an effective
opening gesture, move from there to the closing gesture, and then connect them. There
are a number of reasons for this, but in general terms there is more room to maneuver in
the middle of a counterpoint than there is at either end. In comparison to the middle, the
opening and closing gestures are more constrained and formulaic: there are a number of
rules - ones that requires us to begin and end on a perfect consonance, for instance that limit our choices in the opening and closing measures.
Yet even though we will refrain from composing straight through from the first measure to the last, it still makes sense to begin with the opening: as often as not, if one gets
off to a good start, the rest of the piece will seem to fall into place on its own. Ex. 1.17b
proposes that we begin with G, an octave above the cantus firmus:
Ex. 1.17b
1
2
3
4
5
6
7
8
9
10
11
12
13
Remember, when composing counterpoint above a cantus firmus, we also have the option
of beginning with a perfect 5th, in this case with D. While there is no reason to prefer the
octave to the perfect 5th, neither is there a reason - in this particular instance - for not
beginning on the octave. We have to begin somewhere.
At this point, the question becomes one of how best to continue from G. Since the
well-formedness rules for first species require the exclusive use of consonances, there are
a total of three choices for what note to place in the counterpoint in ms. 2 above A in the
cantus finnus: (1) G can move to F, forming a minor 6th above A in the cantus firmus, (2)
G can leap down to E, forming a perfect 5th with the cantus finnus, or (3) G can leap up
to C, forming a minor 10th with the cantus firmus. Since we cannot use the octave here,
there are no other choices. Without for the moment exploring all three possibilities, we
will go with (I),F, and for this reason: we can continue the motion from G in ms. 1to F in
ms. 2 with motion back to G in ms. 3, forming - in Ex. 1 . 1 7 ~- another minor 6th, this
time above B in the cantus firmus. As what we will learn to recognize in second species as
a neighbor-note motion, G-F'-G forms a simple, compelling stepwise melodic figure with
which to begin the counterpoint. It follows the path of least melodic resistance.
Ex. 1 . 1 7 ~
1
2
3
4
5
6
7
8
9
10
11
12
13
A
8-6-6 counterpoint, which is what we have in the first three measures, is one of the most
common opening gambits in all music, and over the centuries has been used a countless
number of times in a countless number of pieces. In this case, as in numerous others, .it
serves to "couple" the two voices together in parallel 6ths. Ex. 1.17d extends these parallel 6ths as far as B in ms. 5, once again going with the flow, following the path of least
melodic resistance:
Chapter 1
Ex.1.17d
That gives us a sweeping gesture with which to begin the counterpoint. At this point, it
would thus make sense for us to turn our compositional attentions to the conclusion.
Ex. 1.17e suggests F# and G for the last two measures. Because of the prescriptions of
the clausula formalis, there are no other possibilities: the clausula formalis requires us to
end on a perfect 8ve, the G in ms. 13, and requires us to approach the final via semitone
from below, which necessitates the F#- a chromaticism -in ms. 12.
Ex. 1.17e
Here again, however, it would be better to embed a brief motion like this within a larger
melodic gesture, and because stepwise motion into the leading note is the rule, we will
precede it - as in Ex. 1.17f -with G in ms. 11:
Ex. 1.17f
That gives us a G-F#-G neighbor-note motion, a chromatic variant of the melodic figure
with which the counterpoint begins. It also (and again) couples the counterpoint to the
cantus firmus in parallel 6 t h ~because
:
of the long stretch of descending stepwise melodic
motion a t the end of cantus firmus, we can extend these parallel 6ths - this time working
backwards - at least as far as B in ms. 9:
Ex.1.17g
1
2
3
4
5
6
7
8
9
10
11
12
13
A
The smooth, goal-directed conjunct motion, combined with the harmonious parallel Gths
make for a persuasive closing gesture: the opening and closing gestures in this piece are
in fact similar in a number of respects, and even share musical material - we can hear
the closing gesture as a retrograde (more or less) of the opening gesture, as the opening
gesture "in reverse."
Now that we've composed good opening and closing gestures, this would be a good
time to stand back from the music and listen to what we have so far: from here on, the
goal will be to connect these two gestures. As of Ex. 1.17g, there is no one note that acts
as a melodic apex for the counterpoint: B forms the highest note in both the opening and
closing gestures. In order to define a melodic apex, we can extend the closing gesture again working backwards - up to C in ms. 8, as in Ex. 1.17h. C in ms. 8 forms a minor
10th against A in the cantus firmus:
Ex. 1.17h
Working from the other direction, we could of course extend the opening gesture up as
far as C in ms. 6, as Ex. 1.17i does, forming a minor 6th above E in the cantus firmus:
Ex. 1.17i
As the question mark is meant to suggest, however, a C in ms. 6 would defeat the purpose
of our putting a C in ms. 8, which was to form a unique melodic apex for t h e entire counterpoint. For that reason, we will remove the C in ms. 6 and descend
to G.
- as in Ex. 1.17j -
Chapter 1
Ex. 1.17j
1
2
3
4
5
6
7
8
9
10
11
12
13
A
At this point, all that remains for us to do is to fill in one measure: Ex. 1.17k completes
the composition with an E in ms. 7. Now sing ...
Ex. 1.17k
This time (to linger over this one moment a bit longer) there were two possible choices
for ms. 7: A or E. Because of the large leap it forms with C in ms. 8, E would appear to be
the more ambitious choice - it leaves the counterpoint with a large gap to fill in - but
it's without cpestion the better one. To begin with, E opens up the counterpoint in terms
of register, extending the melodic ambitus from C down a minor 6th to E. Without the E,
the melodic range would have extended a mere perfect 5th from C down to F, a narrow
range for a counterpoint this long. But more than that, the presence of E gives rise to certain melodic similarities between the counterpoint and cantus firmus and thus intensifies
the sense of interconnectedness in this particular piece. It allows the counterpoint, like
the cantus firmus, to subdivide a melodic perfect 5th, leaping in this case a major 3rd from
B in ms. 5 down to G in ms. 6 and then a minor 3rd from G in ms. 6 down to E in ms. 7.
And also like the cantus firmus, the counterpoint follows this subdivided leap with a large
leap in the opposite direction, this time a minor 6th from E in ms. 7 to C - the melodic
apex -in ms. 8. This is the most expansive moment in the entire composition: C in ms. 8
rounds off the most intricate melodic gesture in the piece, forms the melodic apex (highest
note) of the entire counterpoint, but also receives the support of the widest harmonic
interval, the minor 10th above the cantus firmus A.
As for the large leap between E in ms. 7 and C in ms. 8,the subsequent melodic continuation resolves and fills it in with motion in the opposite direction. In fact, the note that
completes the process, the penultimate F#,is the leading note in the G mode, and moves
to the final in the next (and last) measure: the entire final gesture can be heard as a response to the minor 6th leap from E in ms. 7 to C in ms. 8. And as usual in a good composition, the counterpoint seems to converge and flow into the final. As the annotations in
Ex. 1.18 are meant to point out, the counterpoint connects E in ms. 7 with the leading
note F# in ms. 12 and then moves to the final G in ms. 13: '
29
Ex. 1.18
Meanwhile, what seems like a separate melodic continuation descends from C in ms. 8
through B and A to the same destination - G - in ms. 13. It is as if there are two separate but concurrent melodic voices in the closing measures of the counterpoint, streams
that flow together into the final, and whose confluence we hear as a single melodic contour. Such composite melodies are common in music of all sorts: musicians sometimes
describe them as polyphonic melodies, melodies comprising more than one simultaneous
melodic voice. As you can probably sense, the musical implications of such melodies are
enormous, for it means that music as simple as two-part, first-species counterpoint can b e
(and often is) heard in terms of more than two voices. In the closing measures of this
particular piece, for instance, there are - counting the cantus firmus - at least three active voices, which the music distributes between two parts. The multivoice nature of these
melodies adds greatly to the complexity of this music.
An additional comment on Ex. 1.18: With regard to the two separate melodic continuations in the final measures of the counterpoint, G in ms. 11 assumes an "in between"
melodic role: it fills in or passes between the "upper voice* A in ms. 10 and the "lower
voice" F# in ms. 12. In the context of Ex. 1.18, G transitions between the two melodic
continuations, as it were, and thus serves to connect them. More - a lot more - on this in
the next chapter.
In this particular piece, the cantus firmus and the counterpoint share an unusual
amount of melodic material: both voices use the same basic melodic figure (down a 3rd,
down a 3rd, up a 4th or 6th) for their most expressive gestures - the two brackets in Ex.
1.17k call our attention to the near simultaneous appearance of this figure in both the
cantus firmus and the counterpoint. That is, the figure is staggered between the two
voices: it begins in the counterpoint one measure before it begins (at a different pitch
level) in the cantus firmus. Too close together, moreover, for us to hear the one in the
cantus firmus as an imitation (or echo) of the one in the counterpoint, or even to recognize
that we're hearing the same or similar figures in both voices - it all passes in an aural
blur. In this sense, the cantus firmus and the counterpoint are out of phase, a factor that
complicates their interaction: amid all the stepwise motion in both voices, the effect of
these overlapping leaps is rather dazzling. But not disorienting. In both voices the melodic
continuation integrates this series of three consecutive leaps into a calmer, more
deliberate motion between two proximate pitches: a virtual melodic connection between
E in ms. 6 and D in ms. 9 - virtual because the two pitches are nonadjacent - draws
these leaps together in the cantus firmus, whereas these leaps are absorbed into the virtual melodic connection between B in ms. 5 and C in ms. 8 in the counterpoint. In the
counterpoint, the fact that it takes so long (four measures) to complete this melodic connection serves to slow down the counterpoint and thus sustain the motion to the melodic
apex.
What general conclusions can we draw from this tutorial? Well, one lesson not to b e
learned is that the average counterpoint and cantus firmus will have this much musical
material in common. Wrong: the amount of musical material the two voices share in this
piece is unusual. In most pieces, the counterpoint and cantus firmus will have little sub-
Chapter B
30
stantive material in common, and there are lots of good pieces in which the contrast
between the two voices is - within limits, of course - maximal, whose constituent gestures are dissimilar, more unalike than alike. It would be wrong to give the impression
that the sharing of musical material is even a musical ideal to strive after, which isn't the
case. Rather, it just happens to be what goes on in this particular piece; one of the great
virtues of species counterpoint is that it gives us concepts and principles with which to
describe what this and other melodies do in precise detail.
One of the conclusions we can draw, on the other hand, is that even though this piece
has a mere 26 notes, few listeners, after working through this tutorial, would describe
species counterpoint as simple or (one hopes) artificial. Species counterpoint is given to
considerable musical complication, for both composer and listener, but also embodies
musical intuitions we all have, intuitions relevant, moreover, to a huge amount of other
music. Perhaps its greatest value is that it allows us to become self-conscious about what
those musical intuitions are and thus to make better use of them.
1.6 Invertible Counterpoint
With one exception - the ill-fated counterpoint in Ex. 1.1371on p. 18 - all the counterpoints we have examined so far were composed to go above a cantus finnus. It is no
less possible, however, to compose a counterpoint to go below. As a matter of fact, most
(but not all) well-formed first-species melodies will work well either above or below the
cantus firmus. As the piece with which we began illustrates: Ex. 1.19a (which reproduces
Ex. 1.1 on p.1) places the counterpoint above the cantus firrnus, while Ex. 1.19b places it
below.
Ex. 1.19a
Ex. 1.19b
Here a counterpoint that begins on the octave above a cantus firmus has been inverted to
begin at the octave below: the term invertible counterpoint refers to a counterpoint that
can be placed both above and below a cantus finnus. Compare the harmonic intervals
between voices in both pieces: a few calculations will confirm that 10ths (and 3 r d ~in
) Ex.
1.19~
become Gths in Ex.1.19b, Gths in Ex.1.19a become lOths in Ex.1.19b, and octaves in
Ex. 1.19a remain octaves in Ex. 1.19b.
As a general rule, when a counterpoint inverts above or below the cantus firmus,
3rds (and 1 0 t h ~will
)
become Gths, Gths will become 3rds (or laths), and octaves
will either remain octaves or become unisons.
The tutorial counterpoint also inverts: Ex. 1.20a (from Ex. 1.17k on p. 28) places the counterpoint above the cantus finnus, while Ex. 1.20b places it below.
Ex.1.20a
Ex. 1.20b
In this case, a counterpoint that begins on the octave above the cantus firmus has been
inverted to begin on the unison (not the octave) below: this time 3rds become Gths, Gths
become 3rds, and octaves become unisons. Note, however, that the one 10th in Ex. 1.20a,
the 10th in ms. 8, becomes a 3rd (rather than a 6th) in Ex. 1.20b. Unlike all the other harmonic intervals in this piece, this one does not invert: because of the registral disposition
of this particular counterpoint, C occurs above the cantus firmus in both Ex. 1.20~and Ex.
1.20b. In Ex.20b, the counterpoint crosses above the cantus firrnus to reach C (the melodic apex) in ms. 8 and then crosses back under below the cantus firmus D in ms. 9.
We can also invert the counterpoint in Ex. 1.20a to begin at the octave (rather than
the unison) below, as in Ex. 1.21~.In order to avoid too low a register for the counterpoint, however, both voices can be transposed up an octave, as in Ex. 1.21b (note the
change from alto to treble clef for the cantus firmus).
Chapter 1
Ex. 1 . 2 1 ~
1
2
3
4
5
6
7
8
9
10
11
12
13
Ex. 1.21b
When composing first-species counterpoint, feel free to transpose the cantus firmus up or
down a n octave in order to maintain the counterpoint within a comfortable range.
As a rule, a counterpoint will invert either above or below if it forms nothing but 3rds,
Gths, and octaves with the cantus firmus, for these intervals remain consonant (and therefore allowable) when turned upside down. A counterpoint that forms a perfect 5th with
the cantus firmus, however, will not invert: when inverted, a consonant perfect 5th becomes a perfect 4th7a dissonance, and is therefore ruled out. Ex. 1.22 demonstrates: Ex.
1.22~
includes a perfect 5th between F and C in ms. 2; when the counterpoint and cantus
f i m u s are inverted, as in Ex. 1.22b, that perfect 5th becomes an unusable (non-wellformed) perfect 4th between C and F.
Ex. 1 . 2 2 ~
Ex. 1.22b
33
In principle, it is no harder to compose first-species counterpoint below a cantus firmus
than above. Because of the well-formedness rule (2 on p. 18) stating that one can begin
on the unison or octave (but not the perfect 5th) below, there are, however, fewer options
for beginning a counterpoint below than one above. But other than that, the same rules
apply, the same concepts and principles remain in force; one follows the same general
procedures whether composing above or below.
1.7 Common Compositional Problems
While the counterpoint in Ex. 1.23 meets all of the well-formedness requirements for
relations between voices - it begins on a perfect Sve, all of the intervals between voices
are consonant, and it concludes with a clausula formalis - few listeners would regard it
as a very good melody.
Ex. 1.23
There are a number of related reasons for this. To begin with, the counterpoint has a narrow range: a puny major 3rd between a low C and high E. Because of its constricted
melodic ambitus, the counterpoint is constrained to repeat the same three notes - C, D,
and E - over and over. That means the tune will be repetitive, and it is, but also that
there will be more than a mere few changes of direction, which there are. The excessive
about faces create the feel of a directionless "zig-zag" and inhibit the formation of larger,
more expansive - more singable - melodic gestures. In addition, the counterpoint repeatedly returns to its upper and lower pitches, which creates the feeling that the counterpoint has artificial and rigid registral boundaries - E seems to form a "ceiling," while
C, a t the other end of the melodic ambitus, forms a "floor." Good melodies, in contrast,
often intimate the existence of registral space both above and below into which the tune
could move, even if it doesn't - there is a sense of registral freedom which the artificial
boundaries in Ex. 1.23, in contrast, negate. In this case, all of these problems combine to
ruin what according to the well-formedness rules of first species otherwise qualifies as a
perfectly "legal" counterpoint.
As a rule, the simpler, more singable, and easier to remember the counterpoint,
the better.
Ex. 1.23 also contains a non-well-formed octave in ms. 4: to occur in the middle of a counterpoint, an octave must form the middle interval in a voice exchange, which the octave in
ms. 4 doesn't.
Chapter 1
Another problem common compositional problem that learners often run into is the
melodic presentation of an augmented 4th or a diminished 5th: besides being dissonant,
this interval is difficult to sing. In Ex. 1.24, for instance, the counterpoint (below) leaps an
augmented 4th from F in ms. 4 to B in ms. 5.
Ex. 1.24
In this case, there are problems in addition to the bad melodic interval: a number of notes
- D in ms. 1, F in ms. 4, and A in ms. 6 - sound like melodic "loose ends," notes abandoned as the counterpoint rises to the final. Indeed, it is because the overall melodic
ascent is so precipitous that these notes are "left hanging" in the first place.
In Ex. 1.25, the counterpoint (now above) outlines a diminished 5th between B in ms.
4 and F i n ms. 7. Even though it subdivides and thus attenuates the dissonance of the interval, the diminished 5th still dominates the counterpoint: its upper and lower notes coincide with changes in melodic direction.
Ex. 1.25
Like the one in Ex. 1.25, the counterpoint in Ex. 1.26 also outlines a melodic diminished
5th, between F in ms. 4 and B in ms. 8. This time, however, the counterpoint fills in the
interval with stepwise motion. Nevertheless, the prominence of F and B as the high and
low notes of the entire counterpoint, and the fact that both notes coincide with changes of
direction, serve to accentuate the melodic dissonance. When composing species counterpoint, avoid direct leaps of an augmented 4th or diminished 5th, but also melodic contours
that outline and emphasize them.
Ex. 1.26
35
At the same time, dissonant melodic contours like these are hard to avoid, in first species
in particular, and lots of listeners don't find them objectionable. A good rule of thumb is
to consider the relative merits (and demerits) of each case and make a decision on that
basis.
Chapter 2
Second Species Counterpoint
2.1 Introduction
On the basis of our discussion of first-species counterpoint, imagine a performance of
Ex. 2.1, then sing the music. As before, use subtle increases and decreases in volume and
vocal intensity to bring out the legato contour of ,each melodic gesture, but also to meld
them into the larger melodic flow of the counterpoint.
Ex. 2.1
In Ex. 2.1, there are two half notes (minims) in the counterpoint for each whole note
(semibreve)in the cantus firmus: this is second-species counterpoint.
A number of obvious differences contrast second with first-species counterpoint: more
notes in the counterpoint, of course, but also the use of dissonances, which did not occur
(and were not allowed) in first species. Ex. 2.2 identifies the dissonances in this counterpoint with arabic numerals. Listen again: none of these dissonances are obtrusive or attract attention to themselves, but rather blend into the melodic flow of the gestures, almost without our noticing them. If anything, these dissonances help to ease the melodic
motion forward from one moment to the next, and that's because - as the brackets in Ex.
2.2 are meant to point out - each of these dissonances occurs within a larger, stepwise
melodic motion.
Ex. 2.2
All of the dissonances in this composition, moreover, fall on the second beat of the measure. As in first species, the strong beats are, without exception, reserved for consonances.
Another difference between second and first-species counterpoint is that melodies in
second species tend to be far more complicated: there are more changes of direction, more
(and larger) leaps than in the average firstbspecies counterpoint, factors that tend to add
complication to melodies. As a result, there are more gestures per counterpoint in second
species, and as the slurs in Ex. 2.3 suggest, some of those gestures contain gestures of
their own:
Ex. 2.3
1
a
a
4
6
fi
7
8
9
10
11
We can hear the first five measures of the counterpoint in Ex. 2.3, for instance, as a single
melodic gesture, which, however, subsumes two smaller subgestures: the first subgesture
extends from D in ms. 1to C in ms. 3; the second subgesture extends from E in ms. 3 to E
in ms. 5. In second species, we can begin to hear melodies in terms of gestures.within gestures.
Yet despite the twists and turns of the counterpoint in Ex. 2.1,it still conforms to our
general criteria for good melodies: the similarities between first and second species are
more profound than their differences. In this particular piece, all the gestures sound
natural and are comfortable to sing, there is a certain confluence of gestures within the
overall contour, the contour has a clear melodic apex, and the entire counterpoint seems
to flow into the final. It also climaxes on or near E in ms. 8, which, as pictured in Ex. 2.4,
serves to connect the melodic apex on F in ms. 4 with the final D in ms. 11,thus guiding
the counterpoint to the cadential goal:
Ex. 2.4
The stepwise melodic continuation from F through E to D, in other words, emerges from a
more convoluted musical surface which, however, it also organizes and controls, one that
moreover includes both conjunct and disjunct motion. Schenker describes the integration
of leaps into prevailing stepwise motion in terms of "melodic fluency," a "flowing songfulness" that gives rise, mixing metaphors, to "a wave-like melodic line," which, "with its ascending and descending curves, maintains an equilibrium in all its individual component
phases."' In most melodies, that is-and not just in species counterpoint -a calmer, more
deliberate, and above all stepwise motion underlies the sometimes extravagant circumlocutions of the music.
Heinrich Schenker, Counterpoint, Vol. 1(1910),tr. John Rothgeb (New York, 1987), p. 94 (translation modified).
38
Chapter 2
In this particular piece, and in marked contrast t o Ex. 1.1 in Chapter 1, most of the
stepwise motion is ascending, while all but one of the leaps are descending: .the descending leaps seem to infuse the counterpoint with energy, which is what powers all the
ascending stepwise motion. In neither case, however, does the correlation of melodic direction with conjunct and disjunct motion represent a norm for its respective species, but
rather just happens to be what these particular melodies do. As in first species, the effect
of a leap in second species is to direct our attention to the immediate musical future, and
to encourage us to imagine, in general terms, a musical continuation that fills in the leap
with stepwise motion. After the leap from E down to G in ms. 8 of Ex. 2.1, that is, we
expect the counterpoint,to change direction and then, eventuaIly, to fill in the gap. And
that's more or less what happens: the counterpoint ascends from G in ms. 8 t o D - the
final - in ms. 11.And because D is the note that completes this process, the leap from E
to G in ms. 8 helps to focus our attention on (and generate melodic momentum toward)
the final.
There is an interesting pattern to the leaps in Ex. 2.1. As the brackets in Ex. 2.4 indicate, the piece begins with a leap of a 4th from D in ms. 1 to A in ms. 2. It then continues
(near the beginning of the second large gesture) with a leap of a 5th from E in ms. 5 down
to A in ms. 6, and (between the second gesture and the third) with a leap of a 6th from E
down to G in ms. 8. Overall, there is a progression from a 4th to a 5th to a 6th, in which
each leap is somewhat more "urgent" than the last, and all three gaps are filled in. The
progression of leaps in Ex.2.5 causes the tune to lean toward the cadence:
Ex.2.5
,
Again, this graduated progression of leaps is not normal for a second-species counterpoint, but rather happens to be what this one does. Species counterpoint gives us language with which to discuss melodic processes in considerable detail and with real precision.
Moreover, a musical process like this one is something we begin to become aware of
only toward the end of the piece: how conscious we are of the melodic process described
in our discussion of Ex. 2.5 above is unclear. Few listeners cognize melodies in terms of
the exact intervals between each note and the next. We barely notice the leap of a 3rd
from C up to E in ms. 3, for instance, even though E is the highest note in the counterpoint up to this point: the ascending 3rd in ms. 3, that is, does not seem to participate in
the progression of descending leaps in Ex.2.5, even though it occurs between the first and
second of those leaps. Nor do the three leaps that do participate in this progression make
equal claims on our attention. As Ex.2.6 below demonstrates, the first of these descending
leaps moves from weak (w)to strong (s), from D on the weak second half-note beat of ms.
1to A on the downbeat of ms. 2, as does the second, which moves from E on the weak
beat of ms. 5 to A on the downbeat of ms. 6. It is because the second weak-to-strong leap
is heard in relation to the first that we hear the second large gesture beginning on E in
ms. 5;in both cases, we hear the leap as an 'upbeat (or anacrusis) to the subsequent ges-
39
ture. The third leap moves, in contrast, from strong to weak, from E on the downbeat of
ms. 8 to G on the weak beat of the same measure.
Ex. 2.6
Because it has a different metrical profile, the third leap is in some sense unanalogous to
the first two in melodic force. And while the first two leaps appear on paper to have
about the same degree of melodic force, the first leap occurs after a half-note rest, which
intensifies the motion from weak beat to strong, giving more of a "bump" to the downbeat
of ms. 2 than the corresponding downbeat in ms. 5. In other words, the metrical definition
of the music (its articulation into strong and weak beats) to some extent obscures the graduated progression of leaps outlined in Ex. 2.5; it emphasizes the unalikeness of the three
leaps.
Most listeners, however, are aware to some degree of moderate to large leaps and expect melodies to compensate for those leaps with more or less stepwise motion in the opposite direction. We listen, that is, for how leaps contribute to the delineation of melodic
gestures and generate tension to which the subsequent melodic continuation will (and
often must) react. As the slurs in Ex. 2.3 suggest, the first two leaps in this series - the
4th from D down to A and the 5th from E down to A -initiate two large melodic gestures,
whereas the third leap - from the 6th E down to G - serves to separ'ate the third gesture
from the second. We do not, that is, hear E on the downbeat of ms. 8 as an anacrusis to
the ensuing melodic gesture, which begins, rather, with the low G. At the same time, we
tend to hear the melodic continuation that follows each of these leaps as responses or
reactions to those leaps, as stepwise fill for the melodic gap. In this sense, the subsequent
stepwise melodic continuation belongs together with the leap that precedes i t and lends it
purpose: the leap of a 4th from D in ms. 1 to A in ms. 2, for instance, occurs within the
extended melodic gesture that begins the piece. And it is in that sense that the slurs in Ex.
2.3 (or for that matter in Ex. 1.4) represent a crude parsing of the counterpoint into melodic gestures. In Ex.2.3, and in most second-species melodies, the leaps both form boundaries that separate gestures and act as the melodic "glue" that binds them together, a
connectedness expressed in performance with legato. Again, gestures do not have precise
boundaries, but rather spill over into one another: gestures overlap. What enables them
to integrate melodic gestures is the preference rule (9 on p. 22) that requires stepwise
motion in the opposite direction following a leap. That rule, which bordering on being a
well-fonnedness rule, ensures that the counterpoint will move back into the registral
space hollowed out by the leap, and that there will be a noticeable overlap in register
between the two musical gestures on either side of the leap.
On paper, and perhaps also in performance (depending, of course, on the performance in question), the last two leaps in the counterpoint would appear to divide the last
five measures of Ex. 2.7 into two repetitive segments: the ascending stepwise 5th from A
in ms. 6 to E in ms. 8 and the ascending stepwise 5th from G i n ms. 8 to the cadential D in
ms. 11:
Chapter 2
Ex. 2.7
S-W
S
-W S
W S
-W S
S
And to some extent this is true: both melodic motions circumscribe stepwise 5ths. It is less
clear, however, whether the second of these melodic 5ths is heard as a sequential repetition of the first: while the first of these melodic motions begins on a strong beat (with A in
ms. 6), the second begins on a weak beat (with G in ms. 8). Moreover, because the two
melodic motions are staggered with respect to the measure, each has a different metrical
profile: the first begins strong-weak-strong, while the second begins weak-strong-weak.
And the whole note in the penultimate measure attenuates the comparison between the
two melodic gestures: the first one covers two and a half measures, the second one three
and a half measures - the second gesture is a full measure longer. It would be difficult,
for these reasons, to hear the second five-note gesture as a straightforward repetition of
the first.
Yet the point is not so much whether or not the third gesture in fact forms a sequential repetition of the first, but that we can, within certain limitations, choose how we hear
the melody, a choice, moreover, that has certain implications for performance. Or rather,
it suggests that how we hear the music and how we perform it are one and the same.
There is a strong inclination in the cultural tradition from which species counterpoint derives to imagine a piece of music and its performance as separate objects, a reflex that we
would do well to resist.
2.2 Basic Concepts
Meter
In second species, the use of two half notes in the counterpoint for each whole note in
the cantus firmus differentiates the music into a series of alternating strong and weak
beats, what musicians refer to as meter - as opposed to the continuous stream of whole
notes in both voices in first species, which, in comparison, is meterless. Ex. 2.8 contrasts a
second-species counterpoint with the cantus firmus in terms of their constituent note
values. It illustrates how the first half note in each measure of the counterpoint coincides
with the onset - or "attack" - of the whole note in the cantus firmus:
Ex.2.8
41
There are, in other words, two simultaneous "percussions" on the first beat of the measure, while on the second beat there is but one. Because of the contrast between two
percussions and one, a metrical accent accrues to the first beat of the measure, in comparison to which the second beat is unaccented. We will refer to accented and unaccented
beats (as we have all along) as strong and weak beats.
It will be worth our while, however, to take a moment and point out that these terms
-strong and weak - are contingent, and far from neutral: the metaphors we use to conceptualize meter (and other musical phenomena) vary considerably across historical time
and among different cultural traditions. When we imagine meter in terms of strong and
weak beats, for instance, we conceptualize them as relative magnitudes, whether of physical force or even political power, to the extent, of course, that there's a difference: strong
beats can be understood to overpower and take precedence over weaker ones. In our own
cultural tradition, and especially with regard to popular music and jazz, we often imagine
accented and unaccented beats as down and upbeats. In this tradition, upbeats are often
heard to relate more to the following downbeat than the previous one, a notion that,
while not untrue, we would like to distance ourselves from, at least in in connection with
species counterpoint: not all upbeats act as pickups to the following downbeats. In German traditions, strong and weak beats are conceptualized in more gravitational terms as
heavy and light beats: beats, in other words, have mass, and therefore weight. In earlier
usage, however, accented and unaccented beats were sometimes imagined as good and
bad, but also as masculine and feminine. It is of course no coincidence that these pairs of
contrastive terms align with and reinforce one another in a certain predictable pattern:
accented beats are strong, heavy, masculine, and therefore good, while unaccented beats
are weak, light, feminine, and therefore bad.
For obvious reasons, it is no longer acceptable to discuss meter in terms of masculine
and feminine beats. Yet we still need language to communicate the intuition that some
beats are more salient (from the Latin salire, to "leap out") than others, and for that reason we will continue to use strong and weak, even though the terms are implicated in a
cultural logic about which we're ambivalent: strong and weak give a more visceral, bodily
sense of the phenomenological qualities of meter than the more neutral downbeat and
upbeat.
More to the point, however - and this is, after all, a digression - these comments on
meter go to show that some of our more intransigent cultural attitudes are ingrained in
our responses to music, even music as simple as species counterpoint. Music matters.
Weak Beats as Means of Transition
In second-species counterpoint, weak beats serve as a means of transition from one
strong beat to the next. In Ex. 2.9, B on the weak beat of ms. 2 passes between A on the
preceding strong beat and C on the strong beat of ms. 3.
Ex. 2.9
Chapter 2
In this particular instance, B forms a dissonance - an augmented 4th - above F in the
cantus firmus, while both A (above F) and C (above E) form consonances. Because it occurs on a weak beat, the dissonance is less noticeable than it would have been had it occurred on a strong beat. As a dissonance, moreover, B is motile (it moves), a condition appropriate to the transitionallconnective nature of the weak beat in second species.
To generalize: weak-beat dissonances can be used in second species to connect strongbeat consonances; all dissonances, moreover, must occur in the context of stepwise motion in the counterpoint. In all cases, the dissonance will form the transitional note in
either (a) a passing-note motion or (b) a neighbor-note motion.' A passing-note motion extends from downbeat to downbeat and consists of stepwise motion all in the same direction, whether up or down. Ex. 2 . 1 0 ~includes a number of dissonant passing-note motions
(a)and one consonant passing-note motion (P):
Ex. 2.10a
A neighbor-note motion also involves stepwise motion from downbeat to downbeat, but in
contrast to passing-note motions, it involves a change in melodic direction, whether up or
down - a change in direction that coincides, moreover, with the transitional note. Ex.
2.10b includes both a dissonant neighbor-note motion (a)and a consonant neighbor-note
motion (P):
Ex. 2.10b
In both passing-note and neighbor-note motions, to summarize, the transitional note can
be either consonant or dissonant.
We should point out that most if not all treatises on species counterpoint forbid dissonant neighbor notes in second species, restricting the use of dissonances in second species
to passing notes. We, however, will allow them, under the above circumstances, and for
the following reasons: (1) since consonant neighbor notes are allowed in second species,
there is nothing at all new or unusual about the concept of a neighbor note per se; (2)
Christopher Bernhard (1628-1692),one' of the first writers to describe the passing note (ca. 1660),
in fact terms it a nota transitus, or transition note. See the Tractatus compositionis augmentatis (after 1657),tr. Walter Hilse, in The Music Forum, Vol. 3, ed. William J. Mitchell and Felix Salzer (New
York, 1973), pp. 78-81.
43
since the rules allow dissonant passing notes on weak beats, weak-beat dissonances (such
as neighbor notes) are common, even routine musical phenomena in second species; (3)
neighbor notes require no additional rules, exemptions, or extra considerations, musical
or otherwise - like passing notes, neighbor notes are both approached and left via stepwise motion; and most importantly, (4) dissonant neighbors sound just fine.
In our discussions of species counterpoint, we often refer to individual notes - rather
than complete melodic motions - as passing notes or neighbor notes, but it is important,
even crucial, to remember that passing-note and neighbor-note motions are "indivisible."
In second species, passing notes and neighbor notes are inseparable, that is, from the
larger, three-note melodic motions that contain them, motions that are heard, moreover,
as single units, whole and complete unto themselves.
There is of course a sense in which a transitional note cannot be heard as a passing
note or neighbor note until the entire three-note motion that subsumes it has reached
completion. In both cases, it is not until the last note in the three-note motion that we
become aware of whether the melodic figure delineates stepwise motion all in the same
direction (passing)or changes directions on the transitional note (neighbor). Yet this does
not mean that we process these melodic motions one note at a time, or that we listen in
each measure with bated breath to hear how each motion will continue. A melodic motion, that is, has the "potential" to pass from the moment of its departure, the moment we
hear the note on the strong beat.3 When it then passes, the motion thus realizes a melodic
potential, one of many, embodied within it from the outset. Even when a potential goes unrealized, it remains a potential, part of what gives that motion its uniqueness: we hear a
given melodic motion, that is, at least in part as a motion that could have continued in a
certain way, but didn't, much like a person who never develops one of his or her talents.
What passes, moreover, is not an individual note or even a series of notes, but rather a
motion, which has to be heard as a single object, even though it occurs over time. As we
listen to melodies, we often sense, at least in general terms, how a certain melodic gesture
will continue before - or rather, as -we hear it, and we do so on the basis of numerous
complex and interacting contextual clues: where the motion occurs within a certain gestural contour, what sort of local melodic motion would best match or enhance the larger
melodic contour, where the gesture occurs within the piece, how imminent the cadence
would appear to be, etc. We cognize melodic gestures, in other words, all at once, and all
of a piece, within a certain limited but open-ended aural radius of melodic possibilities.
Similarly, we often refer to individual notes as dissonances, even though dissonances
are normally defined in terms of intervals -relations between pitches - rather than individual notes. In species counterpoint, we tend to attribute consonance and dissonance properties of relations between pitches - to one but not the other of the notes that participate in that relation. In Ex. 2.11 (on p. 44 below), we thus refer to E on the weak beat of
ms. 4 -rather than the major 9th it forms with D in the cantus firmus - as a dissonance.
Dissonance, in other words, becomes an attribute of notes in the counterpoint, but never
pertains, in contrast, to notes in the cantus firmus. We hear all notes in the cantus firmus,
that is, as consonances, not only on the downbeat of the measure, where cantus firmus
does in fact form consonances with the counterpoint, but throughout their entire durations: in a sense, the cantus firmus sustains the condition of consonance from the strong
beat through the weak beat up to the strong beat of the next measure. For this reason,
the rule that requires us never to leap to or from a dissonance applies to the counterpoint
but not the cantus firmus. In ms. 4 of Ex. 2.11, the counterpoint is constrained to pass via
The notion of "potential"has been adapted from Christopher Hasty, who hears it as an essential
attribute of metrical experience in Meter as Rhythm (Oxford, 1997).
Chapter 2
Ex. 2.11
step from F through a dissonant E to D, while D in the cantus firmus -even though it forms
a dissonance with E in the counterpoint - is free to leap to G in the next measure.
As the above comments are meant to suggest, consonance and dissonance, however
crucial to the practice of species counterpoint, are troubled concepts. It is impossible, at
any rate, to give them definitions that would allow us, in all musical environments and at
all times, to discriminate between them. In our own time, consonances are most often
imagined as "stable" intervals, while dissonances are "unstable," thus requiring continuation. Ideological implications aside - the distinction evinces an unsaid, conservative preference for stable consonances over unstable dissonances, for the status quo over change
- the distinction, while not without credence, sometimes leads to contradictions. In a
second-species composition that cadences on the octave, for instance, the least stable interval will be the major 6th in the penultimate measure: because of the presence of the
leading tone, the penultimate major 6th requires resolution -imagine singing Ex. 2.1 and
stopping on the downbeat of the next-to-last measure! Yet few musicians would be comfortable with the notion that the major 6th, as an unstable interval, is a dissonance. At
least not in the abstract: when sounded out of context, the major 6th, seems quite consonant and euphonious. We could, of course, remain with the definition and judge the major
6th to be dissonant, but that would cause us problems in connection with other major
6 t h we are certain to encounter, some of which will sound altogether stable: compare
the major 6th from F to D in ms. 9 of Ex. 1.1,for instance, with the major 6th from E to Cj
in ms. 10. A more useful position to adopt, however, would be that consonance and dissonance are not inherent properties of an interval but rather are conditions that arise
from the context in which it occurs.
On the Absence of Dissonance in First Species
&
We are now in a position to give a rationale for the definitional absence of dissonance
in first-species counterpoint. Ex. 2.12 inserts a dissonance (an augmented '4th) between
what would otherwise appear to be a succession of first-species consonances:
Ex. 2.12
In the musical culture from which species counterpoint emerged, there tends to be a oneto-one association between dissonances and weak beats; second-species counterpoint, which
45
confines dissonances to weak beats, thus embodies an essential (if contingent) truth about
that culture, but also about musical cognition. Because of that association, we are inclined
to hear the augmented 4th above the second semibreve in Ex. 2.12 as a weak beat in
relation to the consonances that occur (on de facto strong beats) before and after: the
presence of dissonance, in other words, imposes meter on the music and thus converts
first-species counterpoint into what is in essence second-species counterpoint, even though
the two voices maintain a first-species, note-against-note c ~ r r e s ~ o n d e n cOf
e.~
course, other
contextual factors reinforce this hearing: both voices leap to the augmented 4th, accentuating the strong-weak effect, and both voices change directions afterward, moving in
contrary motion to the following minor 6th; as a result, the strong beats 'align with the
two highest notes in the upper voice and the lowest notes in the lower voice. In other
words, the metrical condition of the music is not a mere function of the augmented 4th:
dissonance, that is, never works alone, but rather reinforces other contextual elements
that also contribute to our perception of dissonance.
2.3 Possible Uses of the Second Half Note
Conjunct Motion
In second-species counterpoint, the purpose of the note occurring on the weak beat is to
connect the notes on the adjacent strong beats.5 These connective notes have a number of
uses, the two most important of which involve conjunct motion:
a. Apassing note connects notes a 3rd apart with stepwise motion and thus initiates
motion (whether real or potential) into another register. A passing note, again, can
be either consonant or dissonant, and is inseparable from the three-note melodic
motion in which it is embedded.
b. A neighbor note lies a step above or below the note it neighbors and to which it
returns on the following downbeat: a neighbor extends (or prolongs) the presence
of the note it elaborates from the downbeat of one measure to the next. In this
sense, a neighbor note remains in the same register in which it began, and thus
tends to put the brakes on forward melodic motion. And like the passing note, a
neighbor note forms the transitional note in a three-note melodic motion, and can
be either consonant or dissonant.
Again, Fux does not allow for the dissonant neighbor note in second species. We
can rationalize its inclusion here as follows: (a) second species admits the consonant neighbor note, so there is nothing per se remarkable about the neighbor note
as a melodic figure, and (b) the dissonant neighbor note adheres to the same prinSchenker undertakes a similar demonstration in Counterpoint,Vol. 1,pp. 110-112; Ex. 2.12 adapts
part of his Ex. 151 for our purposes. Schenker, however, frames his argument in harmonic rather
than metrical terms, nor does he make the point that the use of dissonance transforms first species
into second.
We owe this formulation of the idea to Felix Salzer and Carl Schachter, from whom we have also
borrowed the following categorization. See their discussion of the 'Various Functions of the Second
Half Noten in the chapter on second species in Counterpoint in Composition (1969),rev. ed. (New
York, 1989),pp. 43-46.
Chapter 2
46
ciples that govern the dissonant passing note, which are that the dissonance occurs
on a weak beat, and that it occurs in the context of stepwise motion. We can therefore regard the dissonant neighbor note as a conceptual blending of the neighbornote as a melodic figure and the stepwise use of dissonance. The only conceptual
stretch the dissonant neighbor note requires of us is the notion of a change in
melodic direction, which doesn't seem so difficult.
By definition, a neighbor note lies either a semitone or wholetone above or below
the note it neighbors. Ex. 2.10b on p. 42 demonstrates both possibilities: at a, the
neighbor note C lies a semitone above the B it elaborates, while a t P the distance
between D and its upper neighbor E is a wholetone. As a general rule, a neighbor
note sounds best - and will be easiest to manage - when it lies a semitone from
the note it elaborates. In a sense, the semitone eases the neighbor note back from
whence it came: when a neighbor forms a semitone with the main note in the
three-note melodic figure, the path of least melodic resistance is to return to the
main note. Not so when the neighbor note forms a wholetone with the main note,
in which case there is a bit more resistance to returning to the main note.
Because a neighbor-note motion returns to the pitch on which it began, it can, if
not handled carefully, lead to "dead spots" in the counterpoint - zones in which
repeated
~
neighborthe melodic flow stagnates or even grinds to halt. In Ex. 2 . 1 3 ~
note motions result in directionless "slow trills," which move back and forth between the same two notes:
Ex. 2 . 1 3 ~
(As Ex. 2.13a suggests, slow trills tend to cause other problems: because of the
slow trill between A and B, the large leap from B to G in ms. 5 is launched without
preparation, and is awkward to sing, all the more so because G is the apex of the
entire melody.) To avoid melodic stagnation, make sure that a neighbor note
matches the general contour of the melodic gesture in which it occurs. In Ex. 2.13b,
both neighbor-note motions mirror (and round off) the up-down melodic contour of
the larger gesture.
Ex.2.13b
Before going on, now is an opportune moment to discuss the melodic figure that
begins with C in ms. 7 (or thereabouts) and runs at least through the leap from D
down to A in ms. 9, and which includes one of the neighbor motions outlined in Ex.
2.13b above. It is a common melodic figure in second species, often occurring near
the end, before the final ascent into the cadence, as it does here. It features a
descending leap of a perfect 4th approached from above and thus infringes the
preference rule which states that leaps larger than a major 3rd should be preceded as well as followed by motion in the opposite direction: in this case, the leap
from D to A is resolved, but not prepared: E in ms. 8, that is, approaches D on the
downbeat of ms. 9 from above, in the same direction as the leap. We can nevertheless refer to other melodic principles to rationalize this breach of melodic decorum.
To begin with, E in ms. 8 is required in order to fill in the leap from F down to C in
ms. 7. And because E forms the transitional note in a three-note neighbor motion,
there is a sense in which the leap from D down to A in ms. 9 is approached from
below and therefore prepared: if the entire neighbor-note motion forms an expansion (or prolongation) of the D on the downbeat of ms. 8, we can hear that sustained D as having been approached via C from below, as in Ex. 2.14, thus preparing a leap that for all practical purposes descends from D on the downbeat of
ms. 8 to A on the weak beat of ms. 9.
Ex.2.14
Disjunct Motion
In addition to passing-note and neighbor-note motions, the second half note in the measure can have a t least three other melodic functions, all of which involve disjunct motion:
c. The second half note in the measure can sometimes function as a n elided passing note. In Ex.2.15, E in ms. 3 connects C to F and in the process passes over - or
elides - D, which would have formed a dissonance against the cantus firmus. It
Chapter 2
nevertheless consists of melodic motion all in the same direction, and in that sense
simulates passing motion; it gives the impression of a stepwise melodic connection
between downbeats, even though, in this case, there is a leap between C and E.
And because the counterpoint leaps to the elidedpassing note, that note must
form a consonance with the cantus firmus. Remember, all dissonances must be approached and left by step.
Ex.2.15
d. A substitute passing note trades a passing note (whether consonant or dissonant)
for a leap. Two such substitutions occur in Ex. 2.16~.In ms. 7, the transitional,
weak beat C in the melodic motion F-C-D substitutes for E, a passing note, as
shown in Ex. 2.16b. Substitute passing notes provide a little melodic variety but
can also be used to avoid parallel perfect consonances. In Ex. 2.16b, the passing
motion from E to D forms parallel 5ths with the cantus firmus. Use of the substitute passing note in Ex.2.16~eliminates these parallel 5ths, which sound dreadful.
Ex.2 . 1 6 ~
--
Ex.2.16b
e. A subdivided leap divides a larger leap into two smaller intervals. In Ex. 2.17, a
subdivided leap occurs between E on the downbeat of ms. 8'and A on the downbeat of ms. 9. In this case, the second half note in the measure - C - subdivides a
5th into two 3 r d ~ .
Ex.2.17
Here the subdivision extends the leap from downbeat to downbeat and thus slows
it down: the leap takes longer to transpire than it otherwise would. And because
the entire figure has the melodic force of the larger interval, a subdivided leap
must continue with stepwise motion in the opposite direction, which in this case it
does. It should also be "prepared" - approached, that is, from the opposite direction.
In a subdivided leap, two 3rds will combine to form a 5th, or a 3rd and a 4th will
combine to form a 6th - there are no other possibilities if all three notes are to
form consonances with the cantus firmus. Also, since the larger of the two intervals more strongly prescribes motion in the opposite direction, subdivided leaps in
which the smaller interval comes first are easier to pull off. It is preferable, in
other words, to subdivide a 6th into a 3rd and a 4th rather than a 4th and a 3rd. Ex.
2.18 compares the two possible subdivisions of a minor 6th from E to C:
Ex.2.18~
Ex.2.1%
Ex. 2.18a,following the above reasoning, will be easier to hear (and sing) than Ex.
2.18b.
It is also possible to begin a subdivided leap on a weak beat:
Ex. 2.19
Chapter 2
50
As you can hear, the effect is not quite the same: since we hear from strong beat to
strong beat, we are more aware of the stepwise connection from F on the strong
beat of ms. 4 through E on the strong beat of ms. 5 to D on the strong beat of ms. 6
than we are of the perfect 5th that extends from G on the second beat of ms. 4 to C
on the second beat of ms. 5; the fact that the subdivided leap outlines a perfect 5th
makes less of an impression on us. It nevertheless makes for an effective combination of leaps: in this case, the subdivided perfect 5th is both prepared and followed
by stepwise motion in the opposite direction.
2.4 Rules
Almost all the rules governing relations between voices and melodic behavior in first species transfer without further qualification to second species: second species involves less a
new or separate set of rules than the addition of rules to cover situations that never arise
in first species. Those situations pertain, for the most part, to the use of dissonances,
though the rules for second species also relax some of the restrictions that first species
places on perfect consonances. Rules governing the use of dissonances in second species
are for the most part implicit in our concepts of dissonant passing and neighbor-note motions, and will not require much additional discussion.
1. In second-species counterpoint, there are two half notes (minims) in the counterpoint
against each whole note (semibreve) in the cantus fimus. A second-species counterpoint
can begin either together with the cantus firmus or after a half-note rest: Ex. 2.1 on p. 36,
for instance, begins with a half note rest. Both the counterpoint and the cantus firmus,
moreover, must cbnclude with a double whole note (breve), and one can use a whole note
- rather than two half notes -in the penultimate measure.
In second-species counterpoint above the cantus firmus, half notes are rather rare in
the next-to-last measure: the way in which the cantus h u s moves to the final
makes it difficult to maintain half-note motion to the bitter end: more often than
not, one will not be able to form a clausula formalis without using a whole note in
the penultimate measure. Ex. 2 . 2 0 ~adds a counterpoint to the last four measures
of a cantus firmus. If the counterpoint is to conclude with ascending stepwise motion into the leading note but also maintain half-note motion in the penultimate
measure, one almost has to place a D above F on the downbeat of the antepenultimate measure (the measure prior to the penultimate measure):
E x 2.20a
Ex. 2.20b
51
A C (above F)on the downbeat of ms. 9 would form "beaten" parallel 5ths with B
(above E)on the downbeat of ms. 10, nor for the same reason can one use a C (as a
consonant passing note between D and B) on the weak beat of ms. 9. That means a
D almost has to go on the downbeat of the antepenultimate measure. If the leap
from D to A in ms. 9 is to be prepared, however, D must form the goal of ascending
melodic motion in the preceding measure, which means that D serves as the goal
of two separate ascending motions in these measures -B-C-D and B-c#-D -rendering the second but more crucial of these motions (the one to the final in ms. 11)
anticlimactic.
If, on the other hand, we use a whole note C# in the penultimate measure, as in
Ex. 2.20b, the counterpoint becomes much more supple. In this case, it allows us to
use an A on the downbeat of ms. 9 and then pass through B to the whole note C#
in ms. 10. A on the downbeat of ms. 8 in Ex. 2.20b, that is, eliminates the premature D at that same moment in Ex. 2.20a, and because it doesn't require preparation can be approached from either below or above - there are a lot more possibilities, a lot more freedom.
And if there is to be ascending stepwise motion to the leading note in second species below the cantus firmus, half-note motion in the penultimate measure will be
impossible. If we were to attempt to maintain half-note motion in the penultimate
measure, as in Ex. 2.21a, the result would be an unusable (non-well-formed) perfect 4th. Using a whole note in the penultimate measure, in contrast, allows for
smooth, ascending stepwise motion into the leading note, as in Ex. 2.21b.
Ex. 2 . 2 1 ~
Ex.2.21b
Bear no ill will toward whole notes in the penultimate measure: one can waste a
lot of time and energy trying to maintain half-note motion all the way to the very
end, for no real musical gain. It is far more important to compose a good melody
than to have two half notes in the next to last measure.
2. In contrast to first-species counterpoint, both consonances and dissonances are used in
second species, but only consonances can occur on strong beats. In second species, dissonances only occur on weak beats. 2 is a well-formedness rule, to which there are no exceptions.
3. While perfect 5ths and 8ves can occur on strong beats in second species, never place
parallel perfect consonances on the downbeats of consecutive measures. In second-species
counterpoint, we hear from downbeat to downbeat, and because intervals that fall on
strong beats are more noticeable than those that fall on weak ones, the injunction against
parallel perfect consonances extends to consecutive downbeats, even though those inter-
Chapter 2
vals are, strictly speaking, non-adjacent. And because we tend to listen from downbeat to
downbeat, parallel perfect consonances on weak beats are perfectly acceptable, provided
that imperfect consonances fall on the intervening strong beats.
N.B. The rule against parallel perfect consonances still applies to adjacent intervals, both within and between measures.
Thus, in Ex. 2.22a, there are parallel 5ths between consecutive strong beats in ms.
9 and 10, but also parallel 5ths between adjacent intervals, between the weak
beat of ms. 8, that is, and the downbeat of ms. 9. Bad counterpoint.
Ex. 2 . 2 2 ~
Ex. 2.2%
In Ex. 2.22b, on the other hand, there are parallel 5ths between consecutive weak
beats, which, however, sound fine.
Parallel 5 t h and 8ves cannot occur within the measure because the two-againstone half notes in the counterpoint guarantee that relative motion within the measure will be oblique.
'
4. All the rules about beginning and ending on perfect consonances in first species are observed in second species, but in second species one can also use perfect 8ves and unisons
in the middle of a piece. In all cases, however, the constraints governing the use of perfect
consonances in first species remain in effect: no parallel perfect consonances (either on
consecutive beats or - as in 3 above - on the downbeats of consecutive measures), and
no direct perfect consonances. Perfect consonances, that is, must be approached in either
oblique or contrary motion.
Perfect 8ves can occur on either strong beats or weak beats. Unisons, however, can
only be used on weak beats: in a two-voice texture, a unison on a strong beat can
sound as if one of the voices has dropped out. For that reason, unisons are confined
Ex.2.23
53
to weak beats: because of the oblique motion to the unison, both voices "attack"
the doubled pitch a t different moments within the measure, allowing us to remain
aware of two separate melodic voices. Ex. 2.23 (on p. 52 above) includes a unison G
in ms. 8. As with all unisons in second species, it occurs on the weak beat of the
measure, is approached by leap, and forms oblique motion against the cantus firmus. More often than not, a unison will be followed by stepwise motion in the opposite direction.
5. Since imperfect consonances are considered (in the context of species counterpoint) to
be sweeter and more euphonious than perfect consonances, the more imperfect consonances on downbeats, the h l l e r and more sonorous the counterpoint.
This is, of course, more in the nature of a general preference - a maxim - than a
rule, and it should never be allowed to take precedence over the concern to write
good, interesting melodies: counterpoint is all about melodies, and the better the
melodies, the better the counterpoint. In general, be on the guard against the adoption of rules or restrictions that would unnecessarily limit our ability to write good
melodies. For that reason, there are no rules that require a certain ideal mix of imperfect and perfect consonances, or consonances and dissonances, or maxims that
recommend varying the intervals on downbeats. While there are moments in each
composition when one is sensitive to the qualities of intervals between voices, the
crucial point about relations between voices is that those relations never distract
our attention from following both voices as melodies: the counterpoint and cantus
fhnus must agree, allowing them to interact with one another as melodies, and
that's all.
You will also be interested to note that we no longer have need for the rule (7 on
p. 22) that' requires each voice to have a distinctive melodic contour: because the
cantus firmus maintains semibreves against the minims in the counterpoint, there
is no chance of our not hearing them as independent melodies - it is impossible
for them to have the same melodic contours.
6. No ties, repeated notes, or (except for the penultimate and final measures) whole notes,
all of which would disrupt the alternation of strong and weak beats that makes secondspecies counterpoint what it is. Moreover, ties and repeated notes would also tend to give
rise to "dead spots" in the counterpoint, curtailing the forward flow to the final.
The use of dissonance in second-species counterpoint requires the addition of one rule to
those governing melodic behavior in first species:
7. NEVER leap to or from a dissonance! A well-formedness rule, if ever one there
was, though even here there are two exceptions, both of which, however, occur only in
third species, four-against-one counterpoint.
Dissonances in species counterpoint are highly constrained: there are rules in each
species that limit when and how dissonances can occur. In second species, disso-
54
Chapter 2
nances can occur only on weak beats, and only in the context of stepwise motion in
the counterpoint.
Consonances, in contrast, are relatively unconstrained. While there are rules that
limit how one can approach perfect consonances, all consonances - whether perfect or imperfect - can be approached and left freely, by leap as well as by step.
Musical Goals
In second-species counterpoint, the whole idea is to write long, flowing melodic motions across barlines: good melodies combine and integrate a relatively few gestures
into more or less simple contours that sweep to their conclusions. More often than
not, this requires a certain restraint with respect to changes of direction and the use
of leaps, neither of which can be too frequent.
Hence the basic criteria for good melodies in first species remain in force: both the
counterpoint and cantus firmus must harmonize with one another, have distinctive
melodic contours, and flow into their cadential goals.
A Case of is taken Musical Identities
Ex. 2 . 2 4 ~concludes with what appears to be a lower-neighbor motion: as the bracket
points out, the melodic figure involves a change of direction, and the dissonance - the B
in ms. 9 - occurs on a weak beat between two strong-beat consonances, both crucial
characteristics of neighbor-note motions in second species. In this case, however, the two
consonances are formed between two different pitches in the counterpoint: C on the
downbeat of ms. 9 and C# on the downbeat of ms. 10. The melodic figure, in other words,
fails to return to the pitch on which it began: C and C# are non-identical pitches, which
means that B is not their neighbor, and that the melodic gesture as a whole does not form
a neighbor-note motion.
Ex. 2 . 2 4 ~ ~
Even if the melodic figure did form a neighbor-note motion, it would still run afoul of the
rule (13 on p. 23) that prohibits melodic chromaticism in approach to cadence in first speuses a bracket to point out how the downbeats of the last three measures
cies. Ex. 2.24~
give rise to a chromatic motion from through C and C# to D. The rule against melodic
chromaticism also rules out the approach to the cadence in Ex. 2.246, which does not,
however, involve neighbor-note motion in the antepenultiinate measure.
Ex. 2.24b
Apex Repetition
In second species, we can also relax the preference rule (7 on p. 21) that recommends
against repeating the apex note, but only under certain circumstances. An apex can be repeated, that is, when it occurs on a strong beat one time but on a weak beat the other:
Ex. 2.25
F, the highest pitch in the counterpoint, occurs twice in Ex. 2.25, first on the strong beat of
ms.4, then on the weak beat of ms. 7. Because it occurs on a weak beat in ms. 7, we hear
F as an incomplete upper neighbor to the following E, which occurs on the strong beat of
ms. 8: for all intents and purposes, E forms the apex of the melodic arc in these measures.
F, on the other hand, doesn't make as lasting an impression on us on the weak beat of ms.
7 as it does on the strong beat of ms. 4, and for that reason the earlier F is heard as the apex of the entire counterpoint. While repeating the apex is not something we recommend,
this would be an acceptable way of doing it.
A few additional comments before going on. Because of a rather narrow melodic ambitus,
the counterpoint in Ex. 2.25 ends up spending too much time in and around the major 3rd
from C to E, its tessitura. It also changes direction on each note between F on the downbeat of ms. 4 and C on the downbeat of ms. 6, and does in a manner that alternates leaps
with steps, creating a potential sing-song effect. None of these problems spell doom for
the counterpoint, but one has to be on the listen not to exacerbate or call attention to
them in performance. A good idea in this case would be perform the opening gesture so
that it forms a seamless arc that extends to E on the downbeat of ms. 5, making sure the
descent from F to E takes precedence over the final ascent from D to E, thus breaking up
some of the step-leap motion.
Chapter 2
2.5 A Second Species Counterpoint from Gradus a d Parnassum
We used Ex. 2.1 to raise a number of pressing musical issues, but it doesn't represent
the best imaginable second-species counterpoint. In certain respects, the counterpoint is
an odd one, unusual for second species. We have discussed most of the reasons for this
above: the graduated progression of larger and larger leaps, the fact that all those leaps
are descending, the prevalence of ascending stepwise motion in between, and the quasisequential repetition of ascending melodic perfect 5ths over the last six measures.
At the same time, the counterpoint in Ex. 2.1 has its virtues: it is singable, has a rememberable melodic contour, resolves all its melodic tensions, and flows nicely into the
cadence. It thus counts as something of a mixed musical success - a success to be sure,
but not one without qualifications. To be honest, second-species counterpoint is much
harder to compose - or rather, compose well - than first-species counterpoint. To make
matters worse, the cantus firmus in this particular instance is somewhat resistant to good
second-species solutions.
For comparison's sake, Ex. 2.26 downloads one of Fux's attempts to compose a secondspecies counterpoint above this cantus firmus from Gradus ad Parnassum.G
Ex.2.26
Like much of the counterpoint in this treatise, this one is awful: while we have tremendous respect for this treatise as a historical document, not all the lessons it has to teach
are ones we would learn. In this instance, the counterpoint is less wrong - after all, it
follows all the pertinent well-formedness rules - than unmusical. Our goal in this section will be to critique this counterpoint and in the process attempt to improve it.
While there are no "improper" dissonances or parallel perfect consonances in Ex. 2.26,
the counterpoint is problematic nevertheless. In general terms, the counterpoint never
gets going: it generates almost no forward momentum, nor does it flow into the final. It
seems aimless, its movements almost random, wandering here and there without a sense
of clear melodic purpose. This is to a large extent because of its narrow melodic range: for
the most part, the counterpoint in Ex. 2.26 remains within the narrow confines of the
perfect 4th between A and D,. much too restricted an ambitus in second species: because
of the narrow melodic range, the counterpoint is forced to pound on the same three or
four notes over and over. To make matters worse, the counterpoint returns to A and D
again and again, hammering on the lower and upper boundaries of its too-narrow melodic
range. In a sense, the initial leap from A to D in ms. 1serves to define the range within
which the counterpoint will remain and sets the tone - or rather, tones - for all that
follows. In fact, as soon as the counterpoint completes the initial leap of a perfect 4th
J.J. Fux (1660-1741),
Gradus ad Parnassum (17251,partial tr. Alfred Mann in The Study of Counterpoint, ed. W e d Mann (New York, 1943),Fig. 33, p. 45.
57
from A to D, it leaps back a perfect 4th to A on the downbeat of ms. 2: the opening gesture is circular - it goes nowhere. It returns to this perfect 4th, moreover, in ms. 4, where
A leaps up to D, in ms. 6, where D leaps down to A, and in ms. 9, where D again leaps to
A - the repetition is tiresome. And there's more: the closing melodic ascent to the final,
for instance, rises through the same perfect 4th, moving again from A in ms. 9 to D in ms.
11.And even on the rare occasions in which the counterpoint moves beyond the upper
and lower boundaries of its melodic range, it does so in a manner that emphasizes either
A or D: when the counterpoint descends to G in ms. 3, for instance, it immediately
changes directions and goes back to A on the downbeat of ms. 4; and when the counterpoint ascends above D to E in rns. 8, i t returns again to D on the downbeat of the next
ineasure. A and D, D and A, A and D ....
Apart from the overemphasis of its melodic boundaries, the counterpoint seems to fixate on perfects 4th - seven of the ten leaps in this counterpoint are perfect 4ths! Besides
the ones enumerated above, the counterpoint leaps a perfect 4th from C down to G in ms.
3 and from E down to B in ms. 8, the latter of which forms the basis for sequential repetition in the next measure. The two other leaps in this counterpoint, moreover, are both
minor 3rds, and occur between the same two pitches: between D in ms. 4 and B in ms. 5,
and again between B in ms. 8 and D in ms. 9.
Nor are we done at that. Ex. 2.27 outlines another series of melodic repetitions, beginning with the passing motion from A in ms. 2 to C in ms. 3:
Ex.2.27
3
8
3
5
I
r\
B
6
I
,
,
We then hear the same basic passing motion again up a step - see the brackets in Ex.
2.27 - between B in ms. 5 and D in ms. 6 and up another step once more between C in
ms. 7 and E in ms. 8. And each time the counterpoint follows the passing motion with a
descending leap of - guess what? - a perfect 4th.
Of course, the basic reason for the counterpoint's repetitiveness is its narrow melodic
range: within the narrow limits of a perfect 4th, there's nowhere else to go but over the
same five notes, again and again. The large number of leaps and the frequent changes of
direction break the counterpoint up into short segments, which obliterate the sense of
continuous forward motion to the cadence that arises from the large-scale sequence underlined in Ex. 2.27 - the endless succession of melodic snippets make the counterpoint
sound as if i t were pasted together from little bits and pieces of musical material. Whereas good counterpoint seems to converge on and flow into the final, there is little sense of
confluence in the closing measures of this composition. If anything, the tune becomes
rougher, more uneven, and disjunct as it moves toward the cadence. True, the piece does
conclude with a stepwise melodic ascent from A in ms. 9 to D - the final - in ms. 11, but
in terms of creating a forward flow to the cadence, that's too little too late. Not even the
final D imparts a compelling sense of closure to this counterpoint: the tune departs from
and returns to D so many times in this piece that when the final arrives in ms. 11,it no
58
Chapter 2
longer seems like the goal of melodic motion in the preceding measures, but rather
sounds like just another D.
While our criticisms of this counterpoint have been extensive, that does not mean that
we need to discard it, or dismiss it as an example of what not to do. Rather, we can use this
music to discuss the compositional process and explore certain decision-making and
problem-solving skills. When composing species counterpoint, learn to fix problem areas
rather than throwing out whole melodies and starting over from scratch. True, there are
some melodies that can't be repaired, but most can. So, rather than discard this particular
composition, we will listen to hear whether we might not be able to transform it into
more musical counterpoint.
To begin with, the piece needs to get off to a better start. Ex. 2.28~changes D on the
weak beat ofms. 1to G, which replaces the leaps with a lower neighbor motion, resulting
in a much simpler, and much smoother opening figure. In the following six examples, the
brackets above the counterpoint indicate those areas we are making improvements to at
the moment, leaving the rest of the counterpoint as it was in the previous example.
Ex. 2 . 2 8 ~
By reducing the number of times the counterpoint changes direction in the first two and a
half measures &om three (up-down-up) to two (down-up), and by replacing disjunct with
conjunct motion, we can create one long gesture, as opposed to a series of directionless
subgestures put together at random. In this case, changing a single note makes an enormous difference.
Ex. 2.283 revises the next three measures. Here, we have removed all but one of the
leaps in ms. 3 rind 4, and filled in the one that remains - the minor 3rd from B to D with stepwise motion in the opposite direction:
Ex.2.283
At this point, the revised version retains the B-C-D passing motion in ms. 5 and 6 from the
original. Again, fewer leaps, more stepwise motion, and fewer changes of directibn help to
create a sense of larger gestures and a more flowing melody.
Ex. 2.28~continues the ascending stepwise motion that began with B on the downbeat
of ms. 5 in Ex. 2.283 all the way to F on the downbeat of ms. 7 before leaping down a perfect 4th to C in the same measure:
Ex.2.28~
While all the stepwise motion makes for a rather expansive melodic gesture, it also delineates a dissonant melodic contour: in this case, B and F form a diminished (and therefore
dissonant) 5th. Dissonant contours are best avoided, but some of them are unobjectionfor
~ instance, the melodic continuation ameliorates the dissonance of the
able. In Ex. 2 . 2 8 ~
diminished 5th. As the arrows in Ex.2.28d point out, B on the downbeat of ms. 5 resolves
to C on the weak beat of ms. 7, while F on the downbeat of ms. 7 resolves to E on the
downbeat of ms. 8 (both resolutions are virtual). In other words, the dissonant diminished
5th B-F contracts to a consonant C-E major 3rd. Yet even with its dissonant melodic
contour, the revised version is still preferable to the original because of its longer, more
flowing gestures.
Ex.2.28d
Ex. 2.28~2,however, leaves us with a series of three consecutive leaps of a descending
4ths beginning with the one in ms. 7. In order to break up this irksome pattern and create
longer, more flowing gestures, Ex. 2.28e replaces the two half notes in the penultimate
measure of Ex. 2 . 2 8 ~with a whole note, C#,so that instead of beginning on the weak beat
of ms. 9, the closing melodic ascent from A to D now begins on the downbeat of ms. 9.
That eliminates one of the three perfect 4ths and preserves the stepwise motion to the
final.
Chapter 2
Ex. 2.28e
1
2
3
4
5
6
7
8
9
10
11
Ex. 2.2% further replaces A on the weak beat of ms. 8 with D, which eliminates another
perfect 4th and completes the process of filling in the leap from F down a perfect 4th to C
in ms. 7. That extends the long, flowing melodic gesture that begins with D on the
downbeat of ms. 4 to D on the weak beat of ms. 8, a distance of five full measures. Ex.
2.28e also moves A on the weak beat of ms. 9 in Ex. 2.28~ahead to the downbeat. A then
passes through B to C# - now a whole note - in ms. 10, forming a compelling motion to
the final, D, in ms. 11.As opposed to the odd-sounding series of perfect 4ths near the end
of Ex. 2.2&, the closing gesture in Ex.2.2% complements the melodic continuation in the
preceding measures. In fact, the three main gestures meld together so well that it becomes difficult to locate clear internal articulations in this piece, which is about ideal.
One other moment in Ex. 2.2& requires mention, which is the juncture between the
last two gestures in this piece. As can be seen in Ex. 2.28f, the leap from D on the weak
beat of ms. 8 down a perfect 4th to A on the downbeat of ms. 9 is unprepared: it is not
preceded with motion in the opposite direction, but rather approached fi-om the same
direction, above from E. It nevertheless sounds fine, and in fact constitutes a common
melodic figure in second-species counterpoint, one that often occurs near the climax, as
this one does, toward the end of the piece. It is an even more elegant variant - because
the leap occurs from weak to strong - of the melodic figure discussed in connection with
Ex. 2.13b on p. 47 above.
Ex. 2.28f
In summary, the cumulative effect of these revisions has been to smooth out the original counterpoint, reducing the total number of changes in melodic direction, eliminating
a lot of repetition, and evening out what was an excessive amount of disjunct motion.
While not perfect (whatever that would be), the counterpoint in Ex. 2.2& is a good tune
all around: it is singable, has an attractive melodic contour, and flows well into the
cadence. It also retains a number of features and the overall contour of the original - it is
an improved version of the original rather than a whole new counterpoint. In particular,
the opening gesture remains in altered form, as does the general upward melodic drift of
61
the original. Our improved version also preserves a few descending leaps of a perfect 4th,
though the leaps have been shifked and repositioned within the counterpoint, and sound
altogether freer and more natural (to beg a question ... ) than in the original.
An historical postscript: Our efforts to improve Fux's original has an historical (and
ironic) precedent. Fux himself introduces Ex. 2.26 (p. 56) as an improvement - what the
Germans call a Verbesserung - of another counterpoint! His concern in Ex. 2.26 was, of
course, to correct "rule violations" in the original, but as his own revision suggests, good
counterpoint involves much more than the mere elimination of errors.
Chapter 4
Fourth' Species Counterpoint
4.1 Introduction
Perhaps the most remarkable feature of the score for the first-species counterpoint in
Ex.4.1is the signature of one flat, which we have not encountered before:
Ex.4.1
Here the signature indicates that the cantus firmus is a transposition of a D mode cantus
firmus. In this case, a D mode cantus firmus has been transposed down a perfect 5th (or up
a perfect 4th) to begin on G. Because of the one-flat signature, the semitones fall between
E and F and between A and Bb: between 2 and 3, that is, and 8 and ^7 - same as in the D
mode. For this reason, we will need to use "moveable re" solfege, of all things, when singing it:
Ex.4.2
re
do
si
re
sol
fa
mi
do
fa
mi
re
Under moveable re, the semitones remain between mi and fa and between si and do,
which is what we're used to, in all six modes.
In this particular cantus firmus, these modal semitones have a lot to do with how the
separate melodic threads converge on the final. Ex. 4.3 below uses a beam to delineate
the motion from E in ms. 3 (the lowest note in the opening melodic gesture) to F in ms. 8
(the lowest note in the concluding melodic gesture): isolated in register, the ascending motion from 8 and .i guides the cantus firmus toward i - the final G - in ms. ll.The cantus
firmus intersperses the this long-range melodic motion from E to F with repeated motions
from Bb to A: as the other beam in Ex. 4.3 suggests, the descending motion from $ to P also
guides the cantus firmus toward i in the last measure. These two semitonal motions - E-
Fourth Species Counterpoint
63
F below and Bb-A above - narrow in and converge on G from opposite directions, directing the melodic flow of the cantus firmus toward the final.
Ex.4.3
Yet how this particular cantus firmus uses these modal semitones to its own melodic advantage is beside the immediate point: the main reason for the signature in Ex. 4.1 is
pragmatic. In this case, the transposition down a perfect 5th allows us to accommodate
the octave leap in the counterpoint within the normal treble range and renders the notation a bit more manageable, situating both the cantus firmus and counterpoint squarely
in the middle of their respective staves; it maintains both melodies in a singable range.
In more substantive, musical terms, another remarkable aspect of this composition is
the large leap with which the counterpoint begins. As one would expect, the counterpoint
follows this leap with stepwise motion in the opposite direction, filling in - or nearly filling in - the octave. The low D "hangs," but ever so slightly: because it occurred so long
ago, we are only barely aware, once we have arrived on the final G in ms. 11, that the
counterpoint has left opening. D stranded in the lower register. There is also a strong
sense in which the low D and the high D are "the same" .Dythat the lower D has been
"transferred" into a higher register, and that because the high D doesn't hang, the low D
doesn't either.
However one decides to hear the low D, the octave leap lends real melodic purpose to
the counterpoint, which flows into and converges on the final in conformance with our
criteria for good melodies: most of us would agree that this is a good, even exceptional,
first-species composition. In addition t'o its strong sense of forward motion toward and into
the final, it has a clear melodic apex (on the high D in ms. 2), and because of the smaller
leap from F to A later in the counterpoint, there is also a clear sense of climax on or
around G in ms. 9: G fills the gap between F and A and serves as a sort of musical premonition of the final - the leap from F to A allows us to imagine the final well in advance of
its actual arrival in ms. 11. And while the cantus firmus and counterpoint "agree," both
melodies define distinctive melodic contours, each with its own separate melodic apex,
which occurs, moreover, at different times in each voice.
Ex. 4.4 takes Ex.4.1 and shies the entire counterpoint over one half note to the right
in relation to the cantus firmus, which remains the same as before:
Ex.4.4
1
2
3
4
5
6
7
8
9
10
11
Chapter 4
64
While the cantus firmus and counterpoint, as melodies, remain the same (or almost the
same) in both compositions, the change is nevertheless radical: the metrical displacement
in Ex. 4.4 changes the entire musical ethos of Ex. 4.1. In contrast to the euphonious consonances of the first-species counterpoint in Ex. 4.1, the fourth species counterpoint in Ex.
4.4gives rise to dissonances above almost every note in the cantus firmus; the cantus firmus and counterpoint are no longer "unanimous" in their agreement. The dissension within their musical ranks is due, of course, to the fact that the counterpoint has been thrown
out of metrical alignment with the cantus firmus: the counterpoint begins a half measure
later than the cantus firmus and continues to lag behind until the last measure in the
piece. As a result of this metrical displacement, the counterpoint forms dissonances with
the cantus firmus on the strong beat of (in this case) all but a few measures. Whereas the
counterpoint in Ex. 4.1 is consonant and harmonious throughout, the counterpoint in Ex.
4.4 interleaves consonances and dissonances, and places these dissonances a t the most
prominent positions within their respective measures. In contrast to second species, where
dissonances occur on weak beats, dissonances in fourth species occur on strong beats;
fourth species in this sense forms the "reverse" of second species.
'
species
strong beats
weak beats
second
consonance
dissonance or consonance
fourth
dissonance or consonance
consonance
.
At the same time, fourth species is.also similar to first species in that there remains a
basic one-to-one relation between the counterpoint and cantus firmus. With the exception
of FI in the penultimate measure, each note in the counterpoint of Ex. 4.4 is equivalent in
duration to a whole note, but since that whole note no longer fills (or remains within) the
span of a single measure, it must be notated in the form of two half notes tied together
across the barline. Fourth species constitutes a "fractal" species, as it were; it occupies a
conceptual space somewhere "in between" first and second.
In each case, the dissonance that falls on a downbeat in Ex. 4.4 forms a suspension. In
a sense, the notion of a suspension is mirrored in the arc of the tie that carries over suspends - the note across the barline and into the next measure. It thus bridges the distance between a weak beat and the following strong beat, which forms an immeasurable
conceptual gulf or barrier, represented in musical notation as a barline. At the same time,
a dissonant suspension keeps us in suspense, awaiting its resolution to the following consonance: it delays the arrival of the note that, in first species, would otherwise have occurred on the downbeat of the measure. A common denominator in all these suspension
metaphors is an element of tension, both physical and psychological: there is a strong
sense of continual tension in fourth species, a musical tension that arises from the use of
suspensions dissonances.
To be more specific, we can hear each suspension dissonance as a moment of tension,
which the counterpoint then relaxes when it resolves the dissonance to a consonance on
the following weak beat. Ex. 4.5 below uses interval numerals to isolate the first of these
suspension formations within its immediate context. Here the motion of the cantus firmus
to E on the downbeat of ms. 3 "destabilizes" the D tied over above from the previous
measure and renders it mobile. As a suspension, D "presses down" against E in the cantus
Ex.4.5
firmus; downbeats i n fourth species sound heavier, weightier, than downbeats in the other
species. The counterpoint then relieves some (though not all) of this melodic pressure
when it moves down to C on the following weak beat. Hear it? When compared to the unstable, dissonant D, C is consonant and stable.
A fair amount of melodic pressure remains on the counterpoint to descend, of course:
while C resolves the octave leap with stepwise motion in the opposite direction, the leap
has still to be filled in. And some of the downward pressure the suspended D exerted on
the counterpoint transfers to C, even though C doesn't itself become dissonant until the
subsequent strong beat. The recurrent suspensions in the counterpoint, moreover, seem
to infuse the cantus firmus, too, with tension. A suspended dissonance generates a certain
resistance to moving ahead to the next note in the cantus firmus: there is a sense of strain
in its motion to the note below a dissonant suspension. Yet because each dissonance is obligated to resolve down to a consonance on the following weak beat, the net effect of
these suspensions is to urge the music on: dissonant suspensions focus our attention on
the immediate musical future.
Characteristic of fourth species is the manner in which these suspension figures connect together in chains. In ms. 2 of Ex. 4.6, D is tied across the barline, where in ms. 3 it
forms a dissonance above E in the cantus firmus and then resolves to a consonant C. C in
turn is then tied across the barline, where in ms. 4 it forms a dissonance above G in the
cantus firmus and then resolves to a consonant Bb. And so on, from one measure to the
next: between ms. 3 and ms. 7 there are no fewer than five suspensions in a row.
Ex.4.6
In each case, a consonant resolution doubles as a preparation for the next dissonant suspension. This conversion of consonances into dissonances (and vice versa) renews and intensifies the tension from moment to moment and impels both the counterpoint and cantus firmus on to their melodic destinations.
Chapter 4
4.2 Dissonant Suspensions
As in second species, a dissonance in fourth species is not heard as an isolated note,
but is embedded, rather, within a larger, more extensive melodic figure. A suspension is a
melodic figure that coordinates (1)stepwise melodic motion downward with (2) a certain
succession of consonances and dissonances, which aligns, moreover, with (3) a certain metrical succession of strong and weak beats. As a melodic figure, a suspension spans three half
notes and has three component parts, each of which coincides with one of the half notes.
Ex. 4.7 labels each of these three parts: (a)a consonant preparation, which occurs on a
weak beat, (b) the dissonant suspension itself, which occurs on the ensuing strong beat,
and (c) the consonant resolution, which inevitably follows on the successive weak beat.
Ex. 4.7
Here the suspension originates as (a)a consonance (a compound major 6th) on a weak beat
in ms. 2, forms (b) a dissonance (a compound minor 7th) against the cantus firmus on the
strong beat of ms. 3, and then resolves (c)to a consonance (a compound minor 6th) on the
subsequent weak beat. We can summarize the three component parts of a suspension figure in the form of a table:
a.
preparation
consonant
weak
b.
suspension
dissonant
strong
c.
resolution
consonant
weak
In a suspension figure, the same note will, of course, form both the consonant preparation (a) and dissonant suspension (b),a ligature which the tie groups together. At the
same time, however, the musical notation obscures - or at the very least fails to express
- the even more salient melodic connection between the suspension dissonance (b)and
its downward resolution (c)to a consonance. As a melodic figure, a suspension thus comprises two pitches, which break down, however, into three component parts (preparation,
suspension, resolution). Ex. 4.8 below uses an imaginary slur (in quotation marks) to connect the suspension D to the resolution C, but rather than cluttering up the score with
slurs (in addition to all the ties), we will follow the common convention of using a dash to
connect the numbers that designate both the suspension dissonance and the consonant
resolution. As the numbers between staves in Ex. 4.8 indicate, the suspension figure in
Ex. 4.8
1
2
3
4
5
6
7
8
9
10
11
ms. 3 thus forms a 7-6 suspension, where the dash expresses the melodic connection between the dissonant suspension and its downward resolution to a consonance. In this form
of theoretical notation, the preparation is not included: while the interval of preparation
varies from one suspension figure to another, the intervals of suspension and resolution
form but a few recurring patterns, of which the 7-6 configuration is one (more in a moment).
At the risk of becoming repetitive, it is worth re-emphasizing the fact that suspension
dissonances are highly constrained in terms of their melodic resolutions. As with passing
and neighbor notes in second species, one can never leap from a dissonance in fourth species: all dissonant suspensions must resolve via stepwise motion downward. In its most
general form, the rule states that a suspension dissonance must resolve down by step t o
an imperfect consonance. This ensures maximal contrast between the coarseness of the
dissonance and harmoniousness of the imperfect consonance. It also means that some dissonance-consonance configurations will form viable suspensions, but that others will not.
Writing Suspensions above a Cantus Firmus
We will use pairs of numbers to desginate suspenions: one to indicate the dissonant interval of suspension, the other to indicate the consonant interval of resolution. In writing
fourth-species counterpoint above a cantus firmus, there are two possible suspension figures, the 7-6 suspension and the 4-3 suspension. Ex. 4.9 includes one 4-3 suspension (in
ms. 4) and six 7-6 suspensions: as this piece would seem to suggest, 7-6suspensions are
far more numerous - and to be honest, more useful - than 4-3suspensions.
Ex.4.9
As it turns out, 4 and 7 are the only suspension dissonances above a cantus firmus capable of resolving downward to imperfect consonances1 - 7 to 6 and 4 to 3. 9-8 suspensions
are also possible, and some counterpoint manuals permit them, but in a 9-8 suspension
the suspension dissonance resolves downward to a perfect consonance, the effect of which
Chapter 4
68
is rather flat: instead of contrasting a coarse dissonance with a euphonious, sweet sounding imperfect consonance, as the 7-6 suspension in Ex. 4 . 1 0 ~does, the 9-8 suspension in
Ex. 4.10b juxtaposes the asperity of the dissonance with an austere perfect consonance. It
sounds severe: the perfect consonance deprives the resolution of its bliss. Ditto for the 4-5
suspension (below the cantus firmus) in Ex. 4.10c, which Fux,for one, allows. Don't use
them; there's no musical point.
Ex. 4.10a
Ex. 4.10b
Ex. 4.10~
Writing Suspensions below a Cantus firmus
Ex. 4.11 is a model fourth species counterpoint below a cantus firmus. Note that the entire composition relies on a single suspension figure, the 2-3 suspension: the 2-3 suspension and 9-10 suspension, octave equivalents of one another, are the only possible suspension formations in counterpoint below a cantus firmus. 2 and 9 are the only suspension dissonances below a cantus firmus capable of resolving downward to imperfect consonances
-2to3and9to10.
Ex. 4.11
It can be useful to realize that in the case of suspensions below the cantus k u s downward resolution leads to an increase in interval size from suspension to resolution: in a 2-3
suspension, 2 resolves downward to a larger interval, a 3. In contrast, downward resolution of suspensions above a cantus firmus leads to a decrease in interval size from suspension to resolution: in a 7-6 suspension, 7 resolves downward to a 6, while 4-in a 4-3 suspension-resolves downward to a 3, where both consonances are smaller than the dissonances
that precede them. This is because suspensions always resolve down by step, whether
above or below the cantus firmus.
An awareness of the different limitations and possibilities in writing counterpoint
above and below the cantus firmus can help to avoid common mistakes like the one in Ex.
4.12. At first glance -glance being the operative word here - the motion from 7 to 6 appears to form a mundane 7-6 suspension. In this instance, however, 7 ascends to 6, whereas all suspensions must descend to their resolutions: for this reason, 7-6 is not possible
69
as a suspension figure below the cantus firmus; it sounds bad. Remember, counterpoint
above the cantus firmus allows for 7-6 and 4-3 suspensions, while counterpoint below the
cantus firmus allows for 2-3 and 9-10 suspensions.
Ex. 4.12
etc.
Because there are fewer usable suspensions below the cantus firmus, and because there
are more constraints on how the counterpoint can begin in the first place, it is generally
more difficult to write fourth species counterpoint below the cantus firmus than above,
more so, even, than in first or second species.
4.3 Consonant Syncopations
In the last section, we saw (and heard) what happens when a note tied across the barline fonns a dissonance above the cantus firmus-it forms a dissonant suspension. If, however, the tied note forms a consonance above the cantus firmus, as it often does, it forms a
consonant syncopation. Whereas dissonant suspensions are constrained to resolve down
by step, there are almost no constraints on consonant syncopations, which are free to
move by step or by leap in either direction. In ms. 8 in Ex. 4.13, for instance, F forms an
8ve in the counterpoint above F in the cantus firmus: because an 8ve is a consonance, F in
the counterpoint is free to leap to A.
Ex. 4.13
In ms. 3 of Ex. 4.14 below, A in the counterpoint likewise forms a consonance - in this
instance a perfect 5th -with E in the cantus firmus and is thus free to leap to C.
Chapter 4
Ex. 4.14
Ex. 4.14 in fact begins with a series of two consonant syncopations before C in ms. 3 is tied
across the barline, where it forms a suspension dissonance below the cantus firmus D in
ms. 4. Like dissonant suspensions, consonant syncopations can be connected together in
chains, but we can also connect them to dissonant suspensions.
Ex. 4.15 begins with a n even longer series of consonant syncopations. Here D in ms. 1
is tied across the barline, forming a 10th - a consonant syncopation - below F in ms. 2.
D then leaps up to A, a 6th, which is tied across the barline, forming a 5th - another consonant syncopation - below E in ms. 3.
Ex. 4.15
For the next two measures, 5ths and 6ths then alternate, each strong-beat 5th moving to
a 6th on the subsequent weak beat. Like chains of dissonant suspensions, series of consonant syncopations allow for continuous stepwise motion in the counterpoint. In this case,
a series of 5-6 syncopations flows into and merges with a series of 9-10 suspensions, allowing for continuous stepwise motion from A in ms. 2 to D in ms. 7.
It is also possible to reverse the order of succession so that the Gths fall on the downbeats, in which case the results are series of 6-5 syncopations, as in Exs. 4.16b and 4.17b
on p. 71 (which we will further discuss in a moment). In successions of 5ths and Gths, in
other words, either interval can be used to initiate a series of consonant syncopations. And
even though both of these series alternate 5ths and 6th, we do not hear them giving rise
to parallel 5ths, though some more conservative counterpoint manuals do in fact disallow
5-6 syncopations on the grounds that the 5ths occurs on downbeats and thus violate the
prohibition against parallel 5ths on the downbeats of consecutive measures. In fourth
species, however, oblique motion between voices means that these 5ths are never sounded
as simultaneities and therefore make less of an impression qua 5ths on the ear.
Because there are fewer possibilities for suspensions below, 5-6 and 6-5 consonant
syncopations are particularly valuable resources when writing counterpoint below a cantus h u s .
71
Ex.4.16 illustrates how series of 5-6 and 6-5 syncopations operate above a cantus
firmus. In counterpoint above, a 5-6 series generates an ascending melodic line,
while a 6-5series causes the melodic line to descend.
Ex. 4.16a
Ex.4.16b
tc.
Ex.4.17 illustrates how series of 5-6 and 6-5 syncopations operate below a cantus
firmus. Here, the series of 5-6 suspensions generates a descending line, while the
6-5 series causes the line to ascend.
Ex.4.17~
'
Ex.4.17b
5-6 and 6-5series are "reciprocal" in their interactions with the cantus firmus: a 5-6 series
will cause the melodic line to ascend in counterpoint above but to descend in counterpoint
below; a 6-5 series will cause the melodic line to descend in counterpoint above but to
ascend in counterpoint below. Each series, that is, causes melodic motion in the opposite
direction when used on "theother side" of the cantus -us.
It is simpler, though, and better, to be aware of the different uses to which these
series can be put. Because both series create continuous descending stepwise motion, the
6-5 series above and the 5-6 series below can be used to simulate chain,suspensions: Ex.
4.15 above uses the 5-6 series below to simulate chain suspensions in ms. 3 and 4. At the
same .time, the 5-6 series above and the 6-5 series below allow for continuous ascending
stepwise motion, which would otherwise be unobtainable in fourth species. And both the
5-6 and 6-5 series represent a dramatic expansion of compositional resources in counterpoint below, where the possibilities for chain suspensions are more limited than in counterpoint above.
General advice: while dissonant suspensions predominate in fourth species, consonant
syncopations are both useful and good, and should not be overlooked.
Unisons
Unions can also be used in fourth species, but must occur on the strong beat as the
result of a note that has been tied across the barline. Ex. 4.18 demonstrates:
Chapter 4
Ex. 4.18
tc.
4.4 Breaking the Species
As in all species, both voices must end with a double whole note, or breve. In order for
a fourth species counterpoint to end with a breve, the series of syncopations must be interrupted in order to allow the counterpoint to move to the final on the downbeat of the
last measure. This interruption of the syncopations is called breaking the species. Ex. 4.19
demonstrates proper cadence formation for a fourth species counterpoint above a cantus
firmus.
Ex. 4.19
In the penultimate measure of the counterpoint, D has been tied over from the previous
measure, forming a dissonance above E in the cantus firmus. As a suspension dissonance,
D resolves down to C#, as the clausula formalis requires. C# is not tied across the barline,
but moves directly to D on the subsequent downbeat, thus "breaking" the species.
All fourth species compositions must end with a suspension cadence. In counterpoint
above the cantus firmus, the penultimate measure will form a 7-6 suspension, where the
dissonance resolves to the leading note. In counterpoint below the cantus firmus, the
penultimate measure will form a 2-3 (or 9-10)suspension, as in Ex. 4.20, where, again,
Ex. 4.20
73
the dissonance resolves to the leading note. In both cases, the species must be broken in
order to arrive a t the final on the downbeat of the last measure.
When composing fourth species counterpoint, it will sometimes be necessary to break
the species on occasions besides the final cadence. It is often necessary to break the species, for instance, a few measures in advance of the cadence in order to set up the cadence.
Ex. 4.21a breaks the species with C on the downbeat of ms. 7. This interruption in the
succession of tied half notes allows C (a consonance) to leap to F on the subsequent weak
beat in order to initiate the series of 7-6 suspensions that flows to and concludes with the
final in ms. 11.
Ex. 4.21a
It would, of course, have been possible to continue with a 4-3 suspension above A in ms.
7, as in Ex. 4.21b, but in this case that would have "stranded" the counterpoint well below
the final in ms. 10, unable to form the required suspension cadence.
Ex. 4.21b
Even more common, however, are situations like the one in Ex. 4 . 2 2 ~where
~
the species
must be broken to avoid an "unusable dissonance" - here a suspension dissonance that
cannot be resolved t o an imperfect consonance - on the downbeat of tlie next measure.
In ms. 4 of Ex. 4.2%, a tied B resolves down to A and thus completes a 2-3 suspension. If
A were then tied across the barline into ms. 5, however, it would form a perfect 4th below
Ex. 4 . 2 2 ~
Chapter 4
74
D in the cantus firmus, a dissonance that cannot be resolved to an imperfect consonance.
In order to avoid the unusable dissonance, the counterpoint in Ex. 4.223 instead breaks
the species and moves back to B on the downbeat of ms. 5.
Ex.4.22b
1
2
3
4
5
6
7
8
9
10
11
You will have noticed that Ex. 4.22b repeats a note, an A, from strong beat to weak on
the downbeat of ms. 9, breaking the species. While repeated notes were forbidden in first
species (the occasional use of the tie being a possible exception), they can be used in fourth
species in this one situation: when it is used afker breaking the species on the downbeat of
the antepenultimate measure to set up a 2-3 or 7-6 suspension cadence; in either case, the
repeated note will be the final, as it is in Ex. 4.22b. The repeated-note suspension cadence
(for want of a better term) is one of the most common and elegant cadential formulae in
late renaissance vocal music, and though it works best in conjunction with longer cantus
firmi, there is no reason for us not to use it in fourth species.'
Besides breaking the species out of necessity - i.e., to form a suspension cadence or to
avoid an unusable dissonance - it is also possible to break the species simply to improve
the melodic contour of the counterpoint. In effect, breaking the species temporarily turns
fourth species into second species counterpoint, and therefore the rules of second species
will govern the segment in which the counterpoint moves in untied half notes. As a rule,
however, break the species as seldom as possible, and return to the tied half notes of
fourth species as soon as possible, on the very next beat.
Musical Goals
,
The whole idea in fourth species is to write continuous series of dissonant suspensions and to continue the succession of tied half notes without interruption for as
long a s possible. In order to do this, one must often sacrifice other musical desiderata - such as independence of voices - and be more tolerant about what constitutes a good melody: the counterpoint and cantus firmus, that is, may have much
the same musical contour, the occasional note may "hang," some leaps may go unprepared, the melodic apex may be repeated, and so on.
? Pietro Aron (ca. 1480-1545)describes the repeated-note suspension cadence in the Trattato della
natura et cognitione di tutti gli tuoni di canto fgurato (Venice, 1523; facs. repr. New York, 1979),
Chapter 36.
75
A common consequence of chain suspensions is to "couplen the counterpoint and
cantus firmus together, both in terms of melodic contour and melodic direction. Ex. 4 . 2 1 ~
(on p. 73 above) illustrates the point. In this particular piece, the counterpoint reproduces
(or almost reproduces) the contour of the cantus firmus step for step and leap for leap. As
a result, the two voices have almost the exact same melodic contour, and apart from the
metrical displacement of the counterpoint against the cantus firmus and the brief measure in which the species is broken, the voices have almost no melodic independence. Yet
it still qualifies as good fourth-species counterpoint.
Indeed, a good fourth-species counterpoint will often seem to "write itself." Because
we are to write suspensions, and because there are severe constraints on how suspension
dissonances resolve, once a fourth-species counterpoint has been begun, it will sometimes seem as though there is little or no choice about how it will continue, at least not in
the short term. And for all the same reasons, fourth-species melodies tend to consist of
more or less extensive passages of downward stepwise motion, interspersed with the
occasional upward leap: the upward leap returns the counterpoint to a higher register,
allowing for downward stepwise motion - via suspension cadence -into the final.
4.5 Rules
Most of the rules governing fourth species counterpoint are embedded in the principles
that regulate the formation and resolution of dissonant suspensions. And because those
principles coordinate "harmonic" relations with certain "melodic" figures, fourth species
blurs the distinction between rules governing relations between voices and those that
control melodic behavior. The following rules restate and summarize those rules without
further illustration:
1.In fourth species, the counterpoint must begin on a perfect consonance after a halfnote rest. I t then continues in a series of half notes tied over the barline from weak beat
to strong, forming syncopations above the cantus firrnus. Because the cantus firmus consists entirely of whole notes, normal motion between voices in fourth species is oblique.
Both the counterpoint and the cantus firmus, however, must conclude with a double
whole note, or breve, and because the breve must occur on the downbeat of the last measure, the ties must be "broken" after the weak half note in the penultimate measure.
In fourth species counterpoint, the word syncopation has two different but related
uses, which can be a cause for some mild terminological confusion. Syncopation
sometimes describes the rhythmic condition that arises from the metrical displacement of one voice relative to the other; the counterpoint, in this sense, is syncopated against the cantus firmus. At other times, a syncopation refers to a consonance that occurs on a downbeat as a result of a note that has been tied over the
barline from the previous weak beat. The term thus designates an abstract metrical state, on the one hand, and a certain configuration of consonant intervals on
the other.
2. In the counterpoint, a strong beat can form either a consonance or a dissonance with
the cantus firmus. A weak beat, in contrast, must always form a consonance.
Chapter 4
76
3. When a strong beat forms a dissonance, the dissonance must occur within a suspension
figure. A suspension dissonance must be prepared as and resolved to a consonance.
4. All the rules about beginning and ending on perfect consonances in first species are observed in fourth species, and, as in second species, octaves and unisons may occur in the
middle of a piece. Unisons on strong beats, however, can only occur a s the result of a note
that has been tied across the barline.
5. Parallel perfect consonances on consecutive strong or weak beats are allowable as
long as those consonances arise from oblique motion and are mediated by imperfect consonances.
6. It is possible to interrupt the succession of tied half notes in the counterpoint in order
to avoid an unusable (unresolvable) dissonance on a downbeat, to prepare for a suspension cadence, or to improve the melodic cqntour. When it does this, the counterpoint
behaves as if it were in second rather than fourth species and must adhere to second-species rules. One should "break" the species as few times as possible in a given counterpoint.
In general, the rules governing melodic behavior in first and second species remain valid
in fourth species. There are, however, a few important additions:
7. As a rule, suspension dissonances must resolve down by step to imperfect consonances.
This rule narrows the viable suspensions down to the 7-6 and 4-3 above and the 2-3 and
9-10 below. Always resolve dissonant suspensions d o w n by step!
As in first and second species, consonances in fourth species are relatively unconstrained: consonances can be approached and left freely, by step or by leap, and in
either direction.
8. In order to meet both the specific requirements of fourth species and the more general
requirements of the clausula formalis, the penultimate measure should include a 7-6 suspension when the counterpoint is above the cantus firmus, or a 2-3 (or 9-10) suspension
when below. This allows the suspension dissonance to resolve to the leading note on the
weak beat of the penultimate measure before cadencing to the final on the following
downbeat.

A manual for species counterpoint

Transcription

Similar documents

sporting goods stores

A2IM`s LA Licensing Day -‐ October 26, 2011

Mobile location services for your family safety Getting

neil cowley trio - Christopher Holland`s Cosmic Harmony

Meshell Ndegeocello bio

Counterpoint

Out Of Fire - wicked plan

Arvo Pärt - The Choral Tintinnabuli compositions

Improvisation in the Philippine Kulintang:

Archetypal Polysemies in Jocuri II by Sabin Pautza